Instruction Set
a b c d e f g h i j k l m n o p q r s t u v w x y z
The operations of the full instruction set.Instruction Set Table, sortable, filterable, Cores, Registers
abortAtomic - principal op code: abort an in-progress atomic extent; operation encoding group
selector: flow-side operations;
add - This is the basic overflowing integer addition. With 32bit values it defines the Cycle on the Mill architecture.
add1u - Increment by one.
addd - Decimal floating point add in current rounding mode.
addde - Decimal floating point add in current rounding to even.
adddfz - Decimal floating point add in current rounding away from zero.
adddn - Decimal floating point add in current rounding to nearest.
adddp - Decimal floating point add in current rounding towards positive infinity.
adddz - Decimal floating point add in current rounding towards zero.
addf - Floating point add in current rounding mode.
addfe - Floating point add in current rounding to even.
addffz - Floating point add in current rounding away from zero.
addfn - Floating point add in current rounding to nearest.
addfp - Floating point add in current rounding towards positive infinity.
addfz - Floating point add in current rounding towards zero.
addp - Pointer addition. Protects the upper 4 reserved bits, can be scaled for indexing.
adds - Alias for add.
addss - Saturating signed integer addition.
addssv - Alias for addss.
addsv - Alias for add.
addsw - Widening signed integer addition.
addswv - Widening signed integer vector addition.
addsx - Excepting signed integer add.
addsxv - Alias for addsx.
addu - Alias for add.
addus - Saturating unsigned integer addition.
addusv - Alias for addus.
adduv - Alias for add.
adduw - Widening unsigned integer addition.
adduwv - Widening unsigned integer vector addition.
addux - Excepting unsigned integer add.
adduxv - Alias for addux.
all - Returns true if the lowest bit in all vector elements is set.
allocStack - principal op code: allocate more memory in the current stack frame; operation
encoding group selector: flow-side operations;
alternate - Interlace two Vectors.
andl - Bitwise and.
andls - Alias for andl.
andlu - Alias for andl.
any - Returns true if the lowest bit in any of the vector elements is set.
argSize - principal op code: set size of memory argument block for following dynCall;
operation encoding group selector: flow-side operations; possible manifest sizes in
bytes: zero, one, two, or four constant bytes;
BID2d - principal op code: decimal format conversion; arithmetic domain: decimal floating
point; operation encoding group selector: exu-side operations; operation signature
kind: belt operand from ops windo
br - Conditionless branch.
brfl - Branch on false predicate.
brtr - Branch on true predicate.
cachebcl - principal op code: cache control operation; memory sub-operation: Load entire code
range to lowest on-chip cache; operation encoding group selector: flow-side operations;
cachebdl - principal op code: cache control operation; memory sub-operation: Load entire data
range to lowest on-chip cache; operation encoding group selector: flow-side operations;
cached - principal op code: cache control operation; memory sub-operation: duplicate cache
line to new address; operation encoding group selector: flow-side operations; operation
signature kind: belt operand from opsWindow, belt operand from opsWindo
cachee - principal op code: cache control operation; memory sub-operation: evict cache line
to memory; operation encoding group selector: flow-side operations; operation signature
kind: belt operand from opsWindow, belt operand from opsWindo
cachex - principal op code: cache control operation; memory sub-operation: discard cache line
without flush; operation encoding group selector: flow-side operations; operation
signature kind: belt operand from opsWindo
call - The general purpose unconditional abstract call operations you will be using when writing general assembly code.
call0 - principal op code: function call; sense of condition: taken unconditionally; number
of results from a call: no result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
call1 - principal op code: function call; sense of condition: taken unconditionally; number
of results from a call: one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
callfl - principal op code: function call; sense of condition: taken if false; number of
results from a call: any number of results; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
callfl0 - principal op code: function call; sense of condition: taken if false; number of
results from a call: no result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
callfl1 - principal op code: function call; sense of condition: taken if false; number of
results from a call: one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
callfln - principal op code: function call; sense of condition: taken if false; number of
results from a call: more than one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
calln - principal op code: function call; sense of condition: taken unconditionally; number
of results from a call: more than one result; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
calltr - principal op code: function call; sense of condition: taken if true; number of
results from a call: any number of results; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
calltr0 - principal op code: function call; sense of condition: taken if true; number of
results from a call: no result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
calltr1 - principal op code: function call; sense of condition: taken if true; number of
results from a call: one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
calltrn - principal op code: function call; sense of condition: taken if true; number of
results from a call: more than one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
carry - principal op code: carry/borrow gang predicate; arithmetic domain: logical;
operation encoding group selector: exu-side operations;
classifyd - Produce an enum value to classify binary floating points into normal values, subnormals, infinities and NaNs etc.
classifyf - Produce an enum value to classify binary floating points into normal values, subnormals, infinities and NaNs etc.
clear - Clear a single indexed bit.
comp - Bitwise ones complement.
con - Drop an immediate constant on the belt.
conform - principal op code: normalize belt contents to argument list; operation encoding
group selector: flow-side operations; possible manifest sizes in bytes: zero, one, two,
or four constant bytes;
countlfl - Count the zero bits from the left.
countltr - Count the one bits from the left.
countrfl - Count the zero bits from the right.
countrtr - Count the one bits from the right.
d2BID - principal op code: decimal format conversion; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
d2DPD - principal op code: decimal format conversion; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
delta - The delta or distance between two pointers. |x - y|
divd - Decimal floating point division in current rounding mode.
divde - Decimal floating point division in current rounding to even.
divdfz - Decimal floating point division in current rounding away from zero.
divdn - Decimal floating point division in current rounding to nearest.
divdp - Decimal floating point division in current rounding towards positive infinity.
divdz - Decimal floating point division in current rounding towards zero.
divf - Floating point division in current rounding mode.
divfe - Floating point division in current rounding to even.
divffz - Floating point division in current rounding away from zero.
divfn - Floating point division in current rounding to nearest.
divfp - Floating point division in current rounding towards positive infinity.
divfz - Floating point division in current rounding towards zero.
divRems - Signed integer division for quotient and remainder.
divRemu - Unsigned integer division for quotient and remainder.
divs - Signed integer division for quotient.
divu - Unsigned integer division for quotient.
DPD2d - principal op code: decimal format conversion; arithmetic domain: decimal floating
point; operation encoding group selector: exu-side operations; operation signature
kind: belt operand from ops windo
enterAtomic - principal op code: begin a new optimistic atomic extent; operation encoding group
selector: flow-side operations;
eql - Integer equality.
eqld - principal op code: equal; arithmetic domain: decimal floating point; NaN-awareness:
NaN-unaware; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops window, belt operand from ops windo
eqldx - principal op code: equal; arithmetic domain: decimal floating point; NaN-awareness:
NaN-aware; operation encoding group selector: exu-side operations; operation signature
kind: belt operand from ops window, belt operand from ops windo
eqlf - principal op code: equal; arithmetic domain: binary floating point; NaN-awareness:
NaN-unaware; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops window, belt operand from ops windo
eqlfx - principal op code: equal; arithmetic domain: binary floating point; NaN-awareness:
NaN-aware; operation encoding group selector: exu-side operations; operation signature
kind: belt operand from ops window, belt operand from ops windo
eqlp - principal op code: equal; arithmetic domain: pointers; operation encoding group
selector: exu-side operations;
eqls - Alias for eql.
eqlu - Alias for eql.
exitAtomic - principal op code: commit an in-progress atomic extent and return status; operation
encoding group selector: flow-side operations;
exscratchf - principal op code: allocate extended scratchpad space in memory; operation encoding
group selector: flow-side operations; possible manifest sizes in bytes: zero, one, two,
or four constant bytes;
externalized - principal op code: convert a floating-point number from internal to memory format;
arithmetic domain: decimal floating point; operation encoding group selector: flow-side
operations;
externalizef - principal op code: convert a floating-point number from internal to memory format;
arithmetic domain: binary floating point; operation encoding group selector: flow-side
operations;
extract - Extract a vector element to a scalar value operand.
exuArgs - principal op code: additional slot for ganged exu operation; operation encoding
group selector: exu-side operations;
f2fd - Convert decimal floating point into binary floating point in current rounding mode.
f2fde - Convert decimal floating point into binary floating point, rounding to even.
f2fdfz - Convert decimal floating point into binary floating point, rounding away from zero.
f2fdn - Convert decimal floating point into binary floating point, rounding to nearest.
f2fdp - Convert decimal floating point into binary floating point, rounding towards positive infinity.
f2fdz - Convert decimal floating point into binary floating point, rounding towards zero.
f2ff - Convert binary floating point into decimal floating point in current rounding mode.
f2ffe - Convert binary floating point into decimal floating point, rounding to even.
f2fffz - Convert binary floating point into decimal floating point, rounding away from zero.
f2ffn - Convert binary floating point into decimal floating point, rounding to nearest.
f2ffp - Convert binary floating point into decimal floating point, rounding towards positive infinity.
f2ffz - Convert binary floating point into decimal floating point, rounding towards zero.
f2sd - Inexactly convert a decimal floating point value to a signed integer, in current rounding mode and normal modulo overflow.
f2sde - Inexactly convert a decimal floating point value to a signed integer, rounding toward even and normal modulo overflow.
f2sdfz - Inexactly convert a decimal floating point value to a signed integer, rounding away from zero and normal modulo overflow.
f2sdn - Inexactly convert a decimal floating point value to a signed integer, rounding toward nearest and normal modulo overflow.
f2sdp - Inexactly convert a decimal floating point value to a signed integer, rounding toward positive infinity and normal modulo overflow.
f2sds - Inexactly convert a decimal floating point value to a signed integer, in current rounding mode and producing saturating result values.
f2sdse - Inexactly convert a decimal floating point value to a signed integer, rounding toward even and producing saturating result values.
f2sdsfz - Inexactly convert a decimal floating point value to a signed integer, rounding away from zero and producing saturating result values.
f2sdsn - Inexactly convert a decimal floating point value to a signed integer, rounding toward nearest and producing saturating result values.
f2sdsp - Inexactly convert a decimal floating point value to a signed integer, rounding toward positive infinity and producing saturating result values.
f2sdsz - Inexactly convert a decimal floating point value to a signed integer, rounding toward zero and producing saturating result values.
f2sdx - Inexactly convert a decimal floating point value to a signed integer, in current rounding mode and producing NaRs on overflow.
f2sdxe - Inexactly convert a decimal floating point value to a signed integer, rounding toward even and producing NaRs on overflow.
f2sdxfz - Inexactly convert a decimal floating point value to a signed integer, rounding away from zero and producing NaRs on overflow.
f2sdxn - Inexactly convert a decimal floating point value to a signed integer, rounding toward nearest and producing NaRs on overflow.
f2sdxp - Inexactly convert a decimal floating point value to a signed integer, rounding toward positive infinity and producing NaRs on overflow.
f2sdxz - Inexactly convert a decimal floating point value to a signed integer, rounding toward zero and producing NaRs on overflow.
f2sdz - Inexactly convert a decimal floating point value to a signed integer, rounding toward zero and normal modulo overflow.
f2sed - Exactly convert a decimal floating point value to a signed integer, in current rounding mode and normal modulo overflow.
f2sede - Exactly convert a decimal floating point value to a signed integer, rounding toward even and normal modulo overflow.
f2sedfz - Exactly convert a decimal floating point value to a signed integer, rounding away from zero and normal modulo overflow.
f2sedn - Exactly convert a decimal floating point value to a signed integer, rounding toward nearest and normal modulo overflow.
f2sedp - Exactly convert a decimal floating point value to a signed integer, rounding toward positive infinity and normal modulo overflow.
f2seds - Exactly convert a decimal floating point value to a signed integer, in current rounding mode and producing saturating result values.
f2sedse - Exactly convert a decimal floating point value to a signed integer, rounding toward even and producing saturating result values.
f2sedsfz - Exactly convert a decimal floating point value to a signed integer, rounding away from zero and producing saturating result values.
f2sedsn - Exactly convert a decimal floating point value to a signed integer, rounding toward nearest and producing saturating result values.
f2sedsp - Exactly convert a decimal floating point value to a signed integer, rounding toward positive infinity and producing saturating result values.
f2sedsz - Exactly convert a decimal floating point value to a signed integer, rounding toward zero and producing saturating result values.
f2sedx - Exactly convert a decimal floating point value to a signed integer, in current rounding mode and producing NaRs on overflow.
f2sedxe - Exactly convert a decimal floating point value to a signed integer, rounding toward even and producing NaRs on overflow.
f2sedxfz - Exactly convert a decimal floating point value to a signed integer, rounding away from zero and producing NaRs on overflow.
f2sedxn - Exactly convert a decimal floating point value to a signed integer, rounding toward nearest and producing NaRs on overflow.
f2sedxp - Exactly convert a decimal floating point value to a signed integer, rounding toward positive infinity and producing NaRs on overflow.
f2sedxz - Exactly convert a decimal floating point value to a signed integer, rounding toward zero and producing NaRs on overflow.
f2sedz - Exactly convert a decimal floating point value to a signed integer, rounding toward zero and normal modulo overflow.
f2sef - Exactly convert a binary floating point value to a signed integer, in current rounding mode and normal modulo overflow.
f2sefe - Exactly convert a binary floating point value to a signed integer, rounding toward even and normal modulo overflow.
f2seffz - Exactly convert a binary floating point value to a signed integer, rounding away from zero and normal modulo overflow.
f2sefn - Exactly convert a binary floating point value to a signed integer, rounding toward nearest and normal modulo overflow.
f2sefp - Exactly convert a binary floating point value to a signed integer, rounding toward positive infinity and normal modulo overflow.
f2sefs - Exactly convert a binary floating point value to a signed integer, in current rounding mode and producing saturating result values.
f2sefse - Exactly convert a binary floating point value to a signed integer, rounding toward even and producing saturating result values.
f2sefsfz - Exactly convert a binary floating point value to a signed integer, rounding away from zero and producing saturating result values.
f2sefsn - Exactly convert a binary floating point value to a signed integer, rounding toward nearest and producing saturating result values.
f2sefsp - Exactly convert a binary floating point value to a signed integer, rounding toward positive infinity and producing saturating result values.
f2sefsz - Exactly convert a binary floating point value to a signed integer, rounding toward zero and producing saturating result values.
f2sefx - Exactly convert a binary floating point value to a signed integer, in current rounding mode and producing NaRs on overflow.
f2sefxe - Exactly convert a binary floating point value to a signed integer, rounding toward even and producing NaRs on overflow.
f2sefxfz - Exactly convert a binary floating point value to a signed integer, rounding away from zero and producing NaRs on overflow.
f2sefxn - Exactly convert a binary floating point value to a signed integer, rounding toward nearest and producing NaRs on overflow.
f2sefxp - Exactly convert a binary floating point value to a signed integer, rounding toward positive infinity and producing NaRs on overflow.
f2sefxz - Exactly convert a binary floating point value to a signed integer, rounding toward zero and producing NaRs on overflow.
f2sefz - Exactly convert a binary floating point value to a signed integer, rounding toward zero and normal modulo overflow.
f2sf - Inexactly convert a binary floating point value to a signed integer, in current rounding mode and normal modulo overflow.
f2sfe - Inexactly convert a binary floating point value to a signed integer, rounding toward even and normal modulo overflow.
f2sffz - Inexactly convert a binary floating point value to a signed integer, rounding away from zero and normal modulo overflow.
f2sfn - Inexactly convert a binary floating point value to a signed integer, rounding toward nearest and normal modulo overflow.
f2sfp - Inexactly convert a binary floating point value to a signed integer, rounding toward positive infinity and normal modulo overflow.
f2sfs - Inexactly convert a binary floating point value to a signed integer, in current rounding mode and producing saturating result values.
f2sfse - Inexactly convert a binary floating point value to a signed integer, rounding toward even and producing saturating result values.
f2sfsfz - Inexactly convert a binary floating point value to a signed integer, rounding away from zero and producing saturating result values.
f2sfsn - Inexactly convert a binary floating point value to a signed integer, rounding toward nearest and producing saturating result values.
f2sfsp - Inexactly convert a binary floating point value to a signed integer, rounding toward positive infinity and producing saturating result values.
f2sfsz - Inexactly convert a binary floating point value to a signed integer, rounding toward zero and producing saturating result values.
f2sfx - Inexactly convert a binary floating point value to a signed integer, in current rounding mode and producing NaRs on overflow.
f2sfxe - Inexactly convert a binary floating point value to a signed integer, rounding toward even and producing NaRs on overflow.
f2sfxfz - Inexactly convert a binary floating point value to a signed integer, rounding away from zero and producing NaRs on overflow.
f2sfxn - Inexactly convert a binary floating point value to a signed integer, rounding toward nearest and producing NaRs on overflow.
f2sfxp - Inexactly convert a binary floating point value to a signed integer, rounding toward positive infinity and producing NaRs on overflow.
f2sfxz - Inexactly convert a binary floating point value to a signed integer, rounding toward zero and producing NaRs on overflow.
f2sfz - Inexactly convert a binary floating point value to a signed integer, rounding toward zero and normal modulo overflow.
f2ud - Inexactly convert a decimal floating point value to a unsigned integer, in current rounding mode and normal modulo overflow.
f2ude - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward even and normal modulo overflow.
f2udfz - Inexactly convert a decimal floating point value to a unsigned integer, rounding away from zero and normal modulo overflow.
f2udn - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward nearest and normal modulo overflow.
f2udp - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward positive infinity and normal modulo overflow.
f2uds - Inexactly convert a decimal floating point value to a unsigned integer, in current rounding mode and producing saturating result values.
f2udse - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward even and producing saturating result values.
f2udsfz - Inexactly convert a decimal floating point value to a unsigned integer, rounding away from zero and producing saturating result values.
f2udsn - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward nearest and producing saturating result values.
f2udsp - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward positive infinity and producing saturating result values.
f2udsz - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward zero and producing saturating result values.
f2udx - Inexactly convert a decimal floating point value to a unsigned integer, in current rounding mode and producing NaRs on overflow.
f2udxe - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward even and producing NaRs on overflow.
f2udxfz - Inexactly convert a decimal floating point value to a unsigned integer, rounding away from zero and producing NaRs on overflow.
f2udxn - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward nearest and producing NaRs on overflow.
f2udxp - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward positive infinity and producing NaRs on overflow.
f2udxz - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward zero and producing NaRs on overflow.
f2udz - Inexactly convert a decimal floating point value to a unsigned integer, rounding toward zero and normal modulo overflow.
f2ued - Exactly convert a decimal floating point value to a unsigned integer, in current rounding mode and normal modulo overflow.
f2uede - Exactly convert a decimal floating point value to a unsigned integer, rounding toward even and normal modulo overflow.
f2uedfz - Exactly convert a decimal floating point value to a unsigned integer, rounding away from zero and normal modulo overflow.
f2uedn - Exactly convert a decimal floating point value to a unsigned integer, rounding toward nearest and normal modulo overflow.
f2uedp - Exactly convert a decimal floating point value to a unsigned integer, rounding toward positive infinity and normal modulo overflow.
f2ueds - Exactly convert a decimal floating point value to a unsigned integer, in current rounding mode and producing saturating result values.
f2uedse - Exactly convert a decimal floating point value to a unsigned integer, rounding toward even and producing saturating result values.
f2uedsfz - Exactly convert a decimal floating point value to a unsigned integer, rounding away from zero and producing saturating result values.
f2uedsn - Exactly convert a decimal floating point value to a unsigned integer, rounding toward nearest and producing saturating result values.
f2uedsp - Exactly convert a decimal floating point value to a unsigned integer, rounding toward positive infinity and producing saturating result values.
f2uedsz - Exactly convert a decimal floating point value to a unsigned integer, rounding toward zero and producing saturating result values.
f2uedx - Exactly convert a decimal floating point value to a unsigned integer, in current rounding mode and producing NaRs on overflow.
f2uedxe - Exactly convert a decimal floating point value to a unsigned integer, rounding toward even and producing NaRs on overflow.
f2uedxfz - Exactly convert a decimal floating point value to a unsigned integer, rounding away from zero and producing NaRs on overflow.
f2uedxn - Exactly convert a decimal floating point value to a unsigned integer, rounding toward nearest and producing NaRs on overflow.
f2uedxp - Exactly convert a decimal floating point value to a unsigned integer, rounding toward positive infinity and producing NaRs on overflow.
f2uedxz - Exactly convert a decimal floating point value to a unsigned integer, rounding toward zero and producing NaRs on overflow.
f2uedz - Exactly convert a decimal floating point value to a unsigned integer, rounding toward zero and normal modulo overflow.
f2uef - Exactly convert a binary floating point value to a unsigned integer, in current rounding mode and normal modulo overflow.
f2uefe - Exactly convert a binary floating point value to a unsigned integer, rounding toward even and normal modulo overflow.
f2ueffz - Exactly convert a binary floating point value to a unsigned integer, rounding away from zero and normal modulo overflow.
f2uefn - Exactly convert a binary floating point value to a unsigned integer, rounding toward nearest and normal modulo overflow.
f2uefp - Exactly convert a binary floating point value to a unsigned integer, rounding toward positive infinity and normal modulo overflow.
f2uefs - Exactly convert a binary floating point value to a unsigned integer, in current rounding mode and producing saturating result values.
f2uefse - Exactly convert a binary floating point value to a unsigned integer, rounding toward even and producing saturating result values.
f2uefsfz - Exactly convert a binary floating point value to a unsigned integer, rounding away from zero and producing saturating result values.
f2uefsn - Exactly convert a binary floating point value to a unsigned integer, rounding toward nearest and producing saturating result values.
f2uefsp - Exactly convert a binary floating point value to a unsigned integer, rounding toward positive infinity and producing saturating result values.
f2uefsz - Exactly convert a binary floating point value to a unsigned integer, rounding toward zero and producing saturating result values.
f2uefx - Exactly convert a binary floating point value to a unsigned integer, in current rounding mode and producing NaRs on overflow.
f2uefxe - Exactly convert a binary floating point value to a unsigned integer, rounding toward even and producing NaRs on overflow.
f2uefxfz - Exactly convert a binary floating point value to a unsigned integer, rounding away from zero and producing NaRs on overflow.
f2uefxn - Exactly convert a binary floating point value to a unsigned integer, rounding toward nearest and producing NaRs on overflow.
f2uefxp - Exactly convert a binary floating point value to a unsigned integer, rounding toward positive infinity and producing NaRs on overflow.
f2uefxz - Exactly convert a binary floating point value to a unsigned integer, rounding toward zero and producing NaRs on overflow.
f2uefz - Exactly convert a binary floating point value to a unsigned integer, rounding toward zero and normal modulo overflow.
f2uf - Inexactly convert a binary floating point value to a unsigned integer, in current rounding mode and normal modulo overflow.
f2ufe - Inexactly convert a binary floating point value to a unsigned integer, rounding toward even and normal modulo overflow.
f2uffz - Inexactly convert a binary floating point value to a unsigned integer, rounding away from zero and normal modulo overflow.
f2ufn - Inexactly convert a binary floating point value to a unsigned integer, rounding toward nearest and normal modulo overflow.
f2ufp - Inexactly convert a binary floating point value to a unsigned integer, rounding toward positive infinity and normal modulo overflow.
f2ufs - Inexactly convert a binary floating point value to a unsigned integer, in current rounding mode and producing saturating result values.
f2ufse - Inexactly convert a binary floating point value to a unsigned integer, rounding toward even and producing saturating result values.
f2ufsfz - Inexactly convert a binary floating point value to a unsigned integer, rounding away from zero and producing saturating result values.
f2ufsn - Inexactly convert a binary floating point value to a unsigned integer, rounding toward nearest and producing saturating result values.
f2ufsp - Inexactly convert a binary floating point value to a unsigned integer, rounding toward positive infinity and producing saturating result values.
f2ufsz - Inexactly convert a binary floating point value to a unsigned integer, rounding toward zero and producing saturating result values.
f2ufx - Inexactly convert a binary floating point value to a unsigned integer, in current rounding mode and producing NaRs on overflow.
f2ufxe - Inexactly convert a binary floating point value to a unsigned integer, rounding toward even and producing NaRs on overflow.
f2ufxfz - Inexactly convert a binary floating point value to a unsigned integer, rounding away from zero and producing NaRs on overflow.
f2ufxn - Inexactly convert a binary floating point value to a unsigned integer, rounding toward nearest and producing NaRs on overflow.
f2ufxp - Inexactly convert a binary floating point value to a unsigned integer, rounding toward positive infinity and producing NaRs on overflow.
f2ufxz - Inexactly convert a binary floating point value to a unsigned integer, rounding toward zero and producing NaRs on overflow.
f2ufz - Inexactly convert a binary floating point value to a unsigned integer, rounding toward zero and normal modulo overflow.
face - principal op code: fetch-ahead count extension; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
fault - Raise an application fault event.
fill - principal op code: retore spilled operand from extended scratchpad; operation
encoding group selector: flow-side operations; possible manifest sizes in bytes: zero,
one, two, or four constant bytes;
flip - principal op code: bit complement; arithmetic domain: logical; operation encoding
group selector: exu-side operations;
flips - Alias for flip.
flipu - Alias for flip.
flowArgs - principal op code: four-byte constant continuation, conditional predicate, or other
ganged argument; operation encoding group selector: flow-side operations; possible
manifest sizes in bytes: zero, one, two, or four constant bytes; operation signature
kind: manifest constan
fmad - principal op code: fused multiply-add or -add/subtract; arithmetic domain: decimal
floating point; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations;
fmade - principal op code: fused multiply-add or -add/subtract; arithmetic domain: decimal
floating point; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations;
fmadfz - principal op code: fused multiply-add or -add/subtract; arithmetic domain: decimal
floating point; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations;
fmadn - principal op code: fused multiply-add or -add/subtract; arithmetic domain: decimal
floating point; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations;
fmadp - principal op code: fused multiply-add or -add/subtract; arithmetic domain: decimal
floating point; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations;
fmadz - principal op code: fused multiply-add or -add/subtract; arithmetic domain: decimal
floating point; rounding policy: toward zero; operation encoding group selector:
exu-side operations;
fmaf - A floating point fused multiply-add. As usual for those, it yields a higher precision than doing it separately, and is faster too.
fmafe - principal op code: fused multiply-add or -add/subtract; arithmetic domain: binary
floating point; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations;
fmaffz - principal op code: fused multiply-add or -add/subtract; arithmetic domain: binary
floating point; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations;
fmafn - principal op code: fused multiply-add or -add/subtract; arithmetic domain: binary
floating point; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations;
fmafp - principal op code: fused multiply-add or -add/subtract; arithmetic domain: binary
floating point; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations;
fmafz - principal op code: fused multiply-add or -add/subtract; arithmetic domain: binary
floating point; rounding policy: toward zero; operation encoding group selector:
exu-side operations;
geq - principal op code: greater than or equal; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
geqd - principal op code: greater than or equal; arithmetic domain: decimal floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
geqdx - principal op code: greater than or equal; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
geqf - principal op code: greater than or equal; arithmetic domain: binary floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
geqfx - principal op code: greater than or equal; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
geqp - Alias for leqp.
geqs - Alias for gtrs.
gequ - Alias for lequ.
gtr - principal op code: greater than; arithmetic domain: logical; operation encoding
group selector: exu-side operations;
gtrd - principal op code: greater than; arithmetic domain: decimal floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
gtrdx - principal op code: greater than; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
gtrf - principal op code: greater than; arithmetic domain: binary floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
gtrfx - principal op code: greater than; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
gtrp - Alias for lssp.
gtrs - principal op code: greater than; arithmetic domain: signed integer; operation
encoding group selector: exu-side operations;
gtru - principal op code: greater than; arithmetic domain: unsigned integer; operation
encoding group selector: exu-side operations;
imp - principal op code: implies; arithmetic domain: logical; operation encoding group
selector: exu-side operations;
imps - Alias for imp.
impu - Alias for imp.
inject - Replace a vector elements with a different scalar value.
inner - principal op code: enter a loop; sense of condition: taken unconditionally; number
of results from a call: any number of results; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
inner0 - principal op code: enter a loop; sense of condition: taken unconditionally; number
of results from a call: no result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
inner1 - principal op code: enter a loop; sense of condition: taken unconditionally; number
of results from a call: one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
innerfl - principal op code: enter a loop; sense of condition: taken if false; number of
results from a call: any number of results; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
innerfl0 - principal op code: enter a loop; sense of condition: taken if false; number of
results from a call: no result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
innerfl1 - principal op code: enter a loop; sense of condition: taken if false; number of
results from a call: one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
innerfln - principal op code: enter a loop; sense of condition: taken if false; number of
results from a call: more than one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
innern - principal op code: enter a loop; sense of condition: taken unconditionally; number
of results from a call: more than one result; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
innertr - principal op code: enter a loop; sense of condition: taken if true; number of
results from a call: any number of results; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
innertr0 - principal op code: enter a loop; sense of condition: taken if true; number of
results from a call: no result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
innertr1 - principal op code: enter a loop; sense of condition: taken if true; number of
results from a call: one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
innertrn - principal op code: enter a loop; sense of condition: taken if true; number of
results from a call: more than one result; operation encoding group selector: flow-side
operations; possible manifest sizes in bytes: zero, one, two, or four constant bytes;
integerd - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: decimal floating point; rounding policy: use current dynamic
rounding mode; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
integerde - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: decimal floating point; rounding policy: to nearest, ties toward
even adjacent value; operation encoding group selector: exu-side operations; condition
code generator: generate condition codes;
integerdfz - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: decimal floating point; rounding policy: to nearest, ties away from
zero; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerdn - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: decimal floating point; rounding policy: toward negative infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerdp - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: decimal floating point; rounding policy: toward positive infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerdz - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: decimal floating point; rounding policy: toward zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
integered - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: decimal floating point; rounding policy: use current dynamic
rounding mode; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
integerede - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: decimal floating point; rounding policy: to nearest, ties toward
even adjacent value; operation encoding group selector: exu-side operations; condition
code generator: generate condition codes;
integeredfz - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: decimal floating point; rounding policy: to nearest, ties away from
zero; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integeredn - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: decimal floating point; rounding policy: toward negative infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integeredp - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: decimal floating point; rounding policy: toward positive infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integeredz - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: decimal floating point; rounding policy: toward zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
integeref - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: binary floating point; rounding policy: use current dynamic rounding
mode; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerefe - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: binary floating point; rounding policy: to nearest, ties toward even
adjacent value; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
integereffz - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: binary floating point; rounding policy: to nearest, ties away from
zero; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerefn - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: binary floating point; rounding policy: toward negative infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerefp - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: binary floating point; rounding policy: toward positive infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerefz - principal op code: round to integral-valued float; exact result required: exact;
arithmetic domain: binary floating point; rounding policy: toward zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
integerf - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: binary floating point; rounding policy: use current dynamic rounding
mode; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerfe - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: binary floating point; rounding policy: to nearest, ties toward even
adjacent value; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
integerffz - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: binary floating point; rounding policy: to nearest, ties away from
zero; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerfn - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: binary floating point; rounding policy: toward negative infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerfp - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: binary floating point; rounding policy: toward positive infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
integerfz - principal op code: round to integral-valued float; exact result required: inexact;
arithmetic domain: binary floating point; rounding policy: toward zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
internalized - principal op code: convert a floating-point number from memory to internal format;
arithmetic domain: decimal floating point; operation encoding group selector: flow-side
operations;
internalizef - principal op code: convert a floating-point number from memory to internal format;
arithmetic domain: binary floating point; operation encoding group selector: flow-side
operations;
lea - principal op code: load effective address; addressing base: pointer; operation
encoding group selector: flow-side operations; possible manifest sizes in bytes: zero,
one, two, or four constant bytes;
left - principal op code: construct a bool vector with N leading false, or count leading
false in a bool vector; arithmetic domain: logical; operation encoding group selector:
exu-side operations;
leq - principal op code: less than or equal; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
leqd - principal op code: less than or equal; arithmetic domain: decimal floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
leqdx - principal op code: less than or equal; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
leqf - principal op code: less than or equal; arithmetic domain: binary floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
leqfx - principal op code: less than or equal; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
leqp - principal op code: less than or equal; arithmetic domain: pointers; operation
encoding group selector: exu-side operations;
leqs - principal op code: less than or equal; arithmetic domain: signed integer; operation
encoding group selector: exu-side operations;
lequ - principal op code: less than or equal; arithmetic domain: unsigned integer;
operation encoding group selector: exu-side operations;
load - principal op code: load from memory; arithmetic domain: logical; addressing base:
pointer; operation encoding group selector: flow-side operations; possible manifest
sizes in bytes: zero or one constant bytes;
loadd - principal op code: load from memory; arithmetic domain: decimal floating point;
addressing base: register; operation encoding group selector: flow-side operations;
possible manifest sizes in bytes: zero, one, two, or four constant bytes; operation
signature kind: base special register, manifest constant, belt operand from opsWindow,
scale factor for indexes in load/store/lea, data width and scalarity (flo
loadf - principal op code: load from memory; arithmetic domain: binary floating point;
addressing base: register; operation encoding group selector: flow-side operations;
possible manifest sizes in bytes: zero, one, two, or four constant bytes; operation
signature kind: base special register, manifest constant, belt operand from opsWindow,
scale factor for indexes in load/store/lea, data width and scalarity (flow),
morsel-sized manifest constan
logBd - principal op code: extract exponent; arithmetic domain: decimal floating point;
operation encoding group selector: exu-side operations;
logBf - principal op code: extract exponent; arithmetic domain: binary floating point;
operation encoding group selector: exu-side operations;
lss - principal op code: less than; arithmetic domain: logical; operation encoding group
selector: exu-side operations;
lssd - principal op code: less than; arithmetic domain: decimal floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
lssdx - principal op code: less than; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
lssf - principal op code: less than; arithmetic domain: binary floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
lssfx - principal op code: less than; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
lssp - principal op code: less than; arithmetic domain: pointers; operation encoding group
selector: exu-side operations;
lsss - principal op code: less than; arithmetic domain: signed integer; operation encoding
group selector: exu-side operations;
lssu - principal op code: less than; arithmetic domain: unsigned integer; operation
encoding group selector: exu-side operations;
mask - principal op code: bitmask conversion between vector of bool and scalar; arithmetic
domain: logical; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops window, data width and scalarity (ex
maxd - principal op code: maximum; arithmetic domain: decimal floating point; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
maxf - principal op code: maximum; arithmetic domain: binary floating point; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
merge - principal op code: bitwise merge; arithmetic domain: logical; operation encoding
group selector: exu-side operations;
mind - principal op code: minimum; arithmetic domain: decimal floating point; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
minf - principal op code: minimum; arithmetic domain: binary floating point; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mul - principal op code: multiplication; arithmetic domain: logical; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
muld - Decimal floating point multiplication in current rounding mode.
mulde - Decimal floating point multiplication in current rounding to even.
muldfz - Decimal floating point multiplication in current rounding away from zero.
muldn - Decimal floating point multiplication in current rounding to nearest.
muldp - Decimal floating point multiplication in current rounding towards positive infinity.
muldz - Decimal floating point multiplication in current rounding towards zero.
mulf - Floating point multplication in current rounding mode.
mulfe - Floating point multplication in current rounding to even.
mulffz - Floating point multplication in current rounding away from zero.
mulfn - Floating point multplication in current rounding to nearest.
mulfp - Floating point multplication in current rounding towards positive infinity.
mulfz - Floating point multplication in current rounding towards zero.
muls - Alias for mul.
mulsf - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: modulo; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulsfe - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: modulo; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulsffz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: modulo; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulsfn - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: modulo; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfp - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: modulo; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfs - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: saturating; rounding policy: use current dynamic rounding mode; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulsfse - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: saturating; rounding policy: to nearest, ties toward even adjacent value;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
mulsfsfz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: saturating; rounding policy: to nearest, ties away from zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulsfsn - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: saturating; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfsp - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: saturating; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfsz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: saturating; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulsfw - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: widening; rounding policy: use current dynamic rounding mode; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulsfwe - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: widening; rounding policy: to nearest, ties toward even adjacent value;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
mulsfwfz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: widening; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulsfwn - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: widening; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfwp - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: widening; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfwz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: widening; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulsfx - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: excepting; rounding policy: use current dynamic rounding mode; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulsfxe - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: excepting; rounding policy: to nearest, ties toward even adjacent value;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
mulsfxfz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: excepting; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulsfxn - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: excepting; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfxp - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: excepting; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulsfxz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: excepting; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulsfz - principal op code: multiplication; arithmetic domain: signed fixed point; overflow
policy: modulo; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulss - principal op code: multiplication; arithmetic domain: signed integer; overflow
policy: saturating; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
mulssv - Alias for mulss.
mulsv - Alias for mul.
mulsw - principal op code: multiplication; arithmetic domain: signed integer; overflow
policy: widening; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
mulswv - principal op code: multiplication; arithmetic domain: signed integer; overflow
policy: widening; scalar vs. vector data: vector; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
mulsx - principal op code: multiplication; arithmetic domain: signed integer; overflow
policy: excepting; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
mulsxv - Alias for mulsx.
mulu - Alias for mul.
muluf - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: modulo; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulufe - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: modulo; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
muluffz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: modulo; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulufn - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: modulo; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufp - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: modulo; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufs - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: saturating; rounding policy: use current dynamic rounding mode; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulufse - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: saturating; rounding policy: to nearest, ties toward even adjacent value;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
mulufsfz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: saturating; rounding policy: to nearest, ties away from zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulufsn - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: saturating; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufsp - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: saturating; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufsz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: saturating; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulufw - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: widening; rounding policy: use current dynamic rounding mode; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulufwe - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: widening; rounding policy: to nearest, ties toward even adjacent value;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
mulufwfz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: widening; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulufwn - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: widening; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufwp - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: widening; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufwz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: widening; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulufx - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: excepting; rounding policy: use current dynamic rounding mode; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
mulufxe - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: excepting; rounding policy: to nearest, ties toward even adjacent value;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
mulufxfz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: excepting; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
mulufxn - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: excepting; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufxp - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: excepting; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
mulufxz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: excepting; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulufz - principal op code: multiplication; arithmetic domain: unsigned fixed point; overflow
policy: modulo; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops window, bit
numbe
mulus - principal op code: multiplication; arithmetic domain: unsigned integer; overflow
policy: saturating; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
mulusv - Alias for mulus.
muluv - Alias for mul.
muluw - principal op code: multiplication; arithmetic domain: unsigned integer; overflow
policy: widening; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
muluwv - principal op code: multiplication; arithmetic domain: unsigned integer; overflow
policy: widening; scalar vs. vector data: vector; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
mulux - principal op code: multiplication; arithmetic domain: unsigned integer; overflow
policy: excepting; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
muluxv - Alias for mulux.
nand - principal op code: bitwise complement of and; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
nands - Alias for nand.
nandu - Alias for nand.
NaR - principal op code: construct and destruct NaRs; arithmetic domain: logical;
operation encoding group selector: exu-side operations;
narrow - principal op code: narrow scalar to half width; arithmetic domain: logical;
operation encoding group selector: exu-side operations;
narrowd - principal op code: narrow scalar to half width; arithmetic domain: decimal floating
point; rounding policy: use current dynamic rounding mode; operation encoding group
selector: exu-side operations;
narrowde - principal op code: narrow scalar to half width; arithmetic domain: decimal floating
point; rounding policy: to nearest, ties toward even adjacent value; operation encoding
group selector: exu-side operations;
narrowdfz - principal op code: narrow scalar to half width; arithmetic domain: decimal floating
point; rounding policy: to nearest, ties away from zero; operation encoding group
selector: exu-side operations;
narrowdn - principal op code: narrow scalar to half width; arithmetic domain: decimal floating
point; rounding policy: toward negative infinity; operation encoding group selector:
exu-side operations;
narrowdp - principal op code: narrow scalar to half width; arithmetic domain: decimal floating
point; rounding policy: toward positive infinity; operation encoding group selector:
exu-side operations;
narrowdz - principal op code: narrow scalar to half width; arithmetic domain: decimal floating
point; rounding policy: toward zero; operation encoding group selector: exu-side
operations;
narrowf - principal op code: narrow scalar to half width; arithmetic domain: binary floating
point; rounding policy: use current dynamic rounding mode; operation encoding group
selector: exu-side operations;
narrowfe - principal op code: narrow scalar to half width; arithmetic domain: binary floating
point; rounding policy: to nearest, ties toward even adjacent value; operation encoding
group selector: exu-side operations;
narrowffz - principal op code: narrow scalar to half width; arithmetic domain: binary floating
point; rounding policy: to nearest, ties away from zero; operation encoding group
selector: exu-side operations;
narrowfn - principal op code: narrow scalar to half width; arithmetic domain: binary floating
point; rounding policy: toward negative infinity; operation encoding group selector:
exu-side operations;
narrowfp - principal op code: narrow scalar to half width; arithmetic domain: binary floating
point; rounding policy: toward positive infinity; operation encoding group selector:
exu-side operations;
narrowfz - principal op code: narrow scalar to half width; arithmetic domain: binary floating
point; rounding policy: toward zero; operation encoding group selector: exu-side
operations;
narrows - Alias for narrow.
narrowss - principal op code: narrow scalar to half width; arithmetic domain: signed integer;
overflow policy: saturating; operation encoding group selector: exu-side operations;
narrowsx - principal op code: narrow scalar to half width; arithmetic domain: signed integer;
overflow policy: excepting; operation encoding group selector: exu-side operations;
narrowu - Alias for narrow.
narrowus - principal op code: narrow scalar to half width; arithmetic domain: unsigned integer;
overflow policy: saturating; operation encoding group selector: exu-side operations;
narrowux - principal op code: narrow scalar to half width; arithmetic domain: unsigned integer;
overflow policy: excepting; operation encoding group selector: exu-side operations;
neg - principal op code: negate; arithmetic domain: logical; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
negd - principal op code: negate; arithmetic domain: decimal floating point; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
negf - principal op code: negate; arithmetic domain: binary floating point; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
negs - Alias for neg.
negss - principal op code: negate; arithmetic domain: signed integer; overflow policy:
saturating; scalar vs. vector data: scalar; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops windo
negssv - Alias for negss.
negsv - Alias for neg.
negsw - principal op code: negate; arithmetic domain: signed integer; overflow policy:
widening; scalar vs. vector data: scalar; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops windo
negswv - principal op code: negate; arithmetic domain: signed integer; overflow policy:
widening; scalar vs. vector data: vector; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops windo
negsx - principal op code: negate; arithmetic domain: signed integer; overflow policy:
excepting; scalar vs. vector data: scalar; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops windo
negsxv - Alias for negsx.
neq - principal op code: not equal; arithmetic domain: logical; operation encoding group
selector: exu-side operations;
neqd - principal op code: not equal; arithmetic domain: decimal floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
neqdx - principal op code: not equal; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
neqf - principal op code: not equal; arithmetic domain: binary floating point;
NaN-awareness: NaN-unaware; operation encoding group selector: exu-side operations;
neqfx - principal op code: not equal; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
neqp - principal op code: not equal; arithmetic domain: pointers; operation encoding group
selector: exu-side operations;
neqs - Alias for neq.
nequ - Alias for neq.
nextDownd - principal op code: next smaller representable value; arithmetic domain: decimal
floating point; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
nextDownf - principal op code: next smaller representable value; arithmetic domain: binary
floating point; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
nextUpd - principal op code: next larger representable value; arithmetic domain: decimal
floating point; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
nextUpf - principal op code: next larger representable value; arithmetic domain: binary
floating point; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
nimp - principal op code: bitwise complement of implies; arithmetic domain: logical;
operation encoding group selector: exu-side operations;
nimps - Alias for nimp.
nimpu - Alias for nimp.
nope - principal op code: no-operation; arithmetic domain: logical; operation encoding
group selector: exu-side operations;
nopf - principal op code: no-operation; operation encoding group selector: flow-side
operations;
nor - principal op code: bitwise complement of inclusive or; arithmetic domain: logical;
operation encoding group selector: exu-side operations;
norddx - principal op code: unordered relation; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
nordfx - principal op code: unordered relation; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
nors - Alias for nor.
noru - Alias for nor.
notl - principal op code: logical not; arithmetic domain: logical; operation encoding group
selector: exu-side operations;
notls - Alias for notl.
notlu - Alias for notl.
nxor - principal op code: bitwise complement of exclusive or; arithmetic domain: logical;
operation encoding group selector: exu-side operations;
nxors - Alias for nxor.
nxoru - Alias for nxor.
ones - principal op code: count of one-bits; arithmetic domain: logical; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
orddx - principal op code: ordered relation; arithmetic domain: decimal floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
ordfx - principal op code: ordered relation; arithmetic domain: binary floating point;
NaN-awareness: NaN-aware; operation encoding group selector: exu-side operations;
orl - Bitwise or.
orls - Alias for orl.
orlu - Alias for orl.
overflow - principal op code: overflow gang predicate; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
pick - principal op code: pick; arithmetic domain: logical; operation encoding group
selector: exu-side picks;
pickup - principal op code: accept a speculative load; arithmetic domain: logical; operation
encoding group selector: exu-side readers;
quantized - principal op code: left value with right quantum; arithmetic domain: decimal
floating point; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops window, belt operand from ops windo
rd - principal op code: hardware reader; arithmetic domain: logical; operation encoding
group selector: exu-side readers;
rdivd - principal op code: reciprocal divide; arithmetic domain: decimal floating point;
operation encoding group selector: exu-side operations;
rdivf - principal op code: reciprocal divide; arithmetic domain: binary floating point;
operation encoding group selector: exu-side operations;
rdivs - principal op code: reciprocal divide; arithmetic domain: signed integer; operation
encoding group selector: exu-side operations;
rdivu - principal op code: reciprocal divide; arithmetic domain: unsigned integer; operation
encoding group selector: exu-side operations;
recur - principal op code: recurnce update; arithmetic domain: logical; operation encoding
group selector: exu-side picks;
refuse - principal op code: reject a speculative load; arithmetic domain: logical; operation
encoding group selector: exu-side readers;
remd - principal op code: divide for remainder; arithmetic domain: decimal floating point;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
remf - principal op code: divide for remainder; arithmetic domain: binary floating point;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
rems - principal op code: divide for remainder; arithmetic domain: signed integer;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
remu - principal op code: divide for remainder; arithmetic domain: unsigned integer;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
rescue - principal op code: copy arguments to front of belt; operation encoding group
selector: flow-side operations; possible manifest sizes in bytes: zero, one, two, or
four constant bytes;
resolve - principal op code: resolve a pointer as a data reference; arithmetic domain:
logical; operation encoding group selector: exu-side operations; operation signature
kind: belt operand from ops windo
retire - principal op code: minimum retire count pseudo-op, or the widths of the retirees;
operation encoding group selector: flow-side operations; possible manifest sizes in
bytes: zero, one, two, or four constant bytes;
retn - principal op code: return from function; sense of condition: taken unconditionally;
operation encoding group selector: flow-side operations; possible manifest sizes in
bytes: zero, one, two, or four constant bytes;
retnfl - principal op code: return from function; sense of condition: taken if false;
operation encoding group selector: flow-side operations; possible manifest sizes in
bytes: zero, one, two, or four constant bytes;
retntr - principal op code: return from function; sense of condition: taken if true;
operation encoding group selector: flow-side operations; possible manifest sizes in
bytes: zero, one, two, or four constant bytes;
reverse - principal op code: bit reverse; arithmetic domain: logical; operation encoding group
selector: exu-side operations;
reversep - Alias for reverse.
rootd - Decimal floating point square root in current rounding mode.
rootde - Decimal floating point square root in current rounding to even.
rootdfz - Decimal floating point square root in current rounding away from zero.
rootdn - Decimal floating point square root in current rounding to nearest.
rootdp - Decimal floating point square root in current rounding towards positive infinity.
rootdz - Decimal floating point square root in current rounding towards zero.
rootf - Floating point square root in current rounding mode.
rootfe - Floating point square root in current rounding to even.
rootffz - Floating point square root in current rounding away from zero.
rootfn - Floating point square root in current rounding to nearest.
rootfp - Floating point square root in current rounding towards positive infinity.
rootfz - Floating point square root in current rounding towards zero.
roots - principal op code: square root; arithmetic domain: signed integer; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
rootu - principal op code: square root; arithmetic domain: unsigned integer; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
rotate - principal op code: bitwise rotate; arithmetic domain: logical; operation encoding
group selector: exu-side readers;
rotatel - principal op code: bitwise rotate; shift direction: toward greater significance;
arithmetic domain: logical; operation encoding group selector: exu-side operations;
condition code generator: generate condition codes;
rotater - principal op code: bitwise rotate; shift direction: toward lesser significance;
arithmetic domain: logical; operation encoding group selector: exu-side operations;
condition code generator: generate condition codes;
rrootd - Decimal floating point reciprocal square root in current rounding mode.
rrootf - Floating point reciprocal square root in current rounding mode.
rroots - principal op code: reciprocal square root; arithmetic domain: signed integer;
operation encoding group selector: exu-side operations;
rrootu - principal op code: reciprocal square root; arithmetic domain: unsigned integer;
operation encoding group selector: exu-side operations;
s2fd - principal op code: convert signed integer to float; arithmetic domain: decimal
floating point; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
s2fde - principal op code: convert signed integer to float; arithmetic domain: decimal
floating point; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
s2fdfz - principal op code: convert signed integer to float; arithmetic domain: decimal
floating point; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
s2fdn - principal op code: convert signed integer to float; arithmetic domain: decimal
floating point; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
s2fdp - principal op code: convert signed integer to float; arithmetic domain: decimal
floating point; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
s2fdz - principal op code: convert signed integer to float; arithmetic domain: decimal
floating point; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops windo
s2ff - principal op code: convert signed integer to float; arithmetic domain: binary
floating point; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
s2ffe - principal op code: convert signed integer to float; arithmetic domain: binary
floating point; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
s2fffz - principal op code: convert signed integer to float; arithmetic domain: binary
floating point; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
s2ffn - principal op code: convert signed integer to float; arithmetic domain: binary
floating point; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
s2ffp - principal op code: convert signed integer to float; arithmetic domain: binary
floating point; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
s2ffz - principal op code: convert signed integer to float; arithmetic domain: binary
floating point; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops windo
s2u - principal op code: convert signed to unsigned; arithmetic domain: logical; overflow
policy: modulo; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops windo
s2us - principal op code: convert signed to unsigned; arithmetic domain: logical; overflow
policy: saturating; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops windo
s2ux - principal op code: convert signed to unsigned; arithmetic domain: logical; overflow
policy: excepting; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops windo
sameQuantumd - principal op code: same exponent; arithmetic domain: decimal floating point;
operation encoding group selector: exu-side operations;
scaleBd - principal op code: scale exponent; arithmetic domain: decimal floating point;
rounding policy: use current dynamic rounding mode; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
scaleBde - principal op code: scale exponent; arithmetic domain: decimal floating point;
rounding policy: to nearest, ties toward even adjacent value; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
scaleBdfz - principal op code: scale exponent; arithmetic domain: decimal floating point;
rounding policy: to nearest, ties away from zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
scaleBdn - principal op code: scale exponent; arithmetic domain: decimal floating point;
rounding policy: toward negative infinity; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
scaleBdp - principal op code: scale exponent; arithmetic domain: decimal floating point;
rounding policy: toward positive infinity; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
scaleBdz - principal op code: scale exponent; arithmetic domain: decimal floating point;
rounding policy: toward zero; operation encoding group selector: exu-side operations;
condition code generator: generate condition codes;
scaleBf - principal op code: scale exponent; arithmetic domain: binary floating point;
rounding policy: use current dynamic rounding mode; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
scaleBfe - principal op code: scale exponent; arithmetic domain: binary floating point;
rounding policy: to nearest, ties toward even adjacent value; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
scaleBffz - principal op code: scale exponent; arithmetic domain: binary floating point;
rounding policy: to nearest, ties away from zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
scaleBfn - principal op code: scale exponent; arithmetic domain: binary floating point;
rounding policy: toward negative infinity; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
scaleBfp - principal op code: scale exponent; arithmetic domain: binary floating point;
rounding policy: toward positive infinity; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
scaleBfz - principal op code: scale exponent; arithmetic domain: binary floating point;
rounding policy: toward zero; operation encoding group selector: exu-side operations;
condition code generator: generate condition codes;
scratchf - principal op code: allocate scratchpad frame; arithmetic domain: logical; operation
encoding group selector: exu-side readers;
set - Set a single indexed bit.
shiftl - Bitwise left shift.
shiftls - Alias for shiftl.
shiftlss - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: signed integer; overflow policy: saturating; scalar vs. vector data:
scalar; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftlssv - Alias for shiftlss.
shiftlsv - Alias for shiftl.
shiftlsw - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: signed integer; overflow policy: widening; scalar vs. vector data:
scalar; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftlswv - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: signed integer; overflow policy: widening; scalar vs. vector data:
vector; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftlsx - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: signed integer; overflow policy: excepting; scalar vs. vector data:
scalar; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftlsxv - Alias for shiftlsx.
shiftlu - Alias for shiftl.
shiftlus - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: unsigned integer; overflow policy: saturating; scalar vs. vector
data: scalar; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftlusv - Alias for shiftlus.
shiftluv - Alias for shiftl.
shiftluw - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: unsigned integer; overflow policy: widening; scalar vs. vector data:
scalar; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftluwv - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: unsigned integer; overflow policy: widening; scalar vs. vector data:
vector; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftlux - principal op code: bitwise shift; shift direction: toward greater significance;
arithmetic domain: unsigned integer; overflow policy: excepting; scalar vs. vector
data: scalar; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftluxv - Alias for shiftlux.
shiftrs - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed integer; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, bit numbe
shiftrsf - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed fixed point; rounding policy: use current dynamic rounding
mode; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrsfe - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed fixed point; rounding policy: to nearest, ties toward even
adjacent value; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftrsffz - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed fixed point; rounding policy: to nearest, ties away from
zero; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrsfn - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed fixed point; rounding policy: toward negative infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrsfp - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed fixed point; rounding policy: toward positive infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrsfz - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: signed fixed point; rounding policy: toward zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
shiftru - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned integer; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
shiftruf - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned fixed point; rounding policy: use current dynamic rounding
mode; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrufe - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned fixed point; rounding policy: to nearest, ties toward even
adjacent value; operation encoding group selector: exu-side operations; condition code
generator: generate condition codes;
shiftruffz - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned fixed point; rounding policy: to nearest, ties away from
zero; operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrufn - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned fixed point; rounding policy: toward negative infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrufp - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned fixed point; rounding policy: toward positive infinity;
operation encoding group selector: exu-side operations; condition code generator:
generate condition codes;
shiftrufz - principal op code: bitwise shift; shift direction: toward lesser significance;
arithmetic domain: unsigned fixed point; rounding policy: toward zero; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
shuffle - Create a new vector where the elements are the values of the first operand vector and their respective positions in the second operand vector.
smeari - principal op code: smear first one to end of vector; arithmetic domain: logical;
whether inclusive or exclusive of boundary: inclusive; operation encoding group
selector: exu-side operations;
smearx - principal op code: smear first one to end of vector; arithmetic domain: logical;
whether inclusive or exclusive of boundary: exclusive; operation encoding group
selector: exu-side operations;
spill - principal op code: spill operand to extended scratchpad; operation encoding group
selector: flow-side operations; possible manifest sizes in bytes: zero, one, two, or
four constant bytes;
splat - principal op code: replicate scalar to vector; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
stackf - principal op code: allocate stack frame; operation encoding group selector:
flow-side operations; possible manifest sizes in bytes: zero, one, two, or four
constant bytes;
store - principal op code: store to memory; arithmetic domain: logical; addressing base:
register; operation encoding group selector: flow-side operations; possible manifest
sizes in bytes: zero, one, two, or four constant bytes;
stored - principal op code: store to memory; arithmetic domain: decimal floating point;
operation encoding group selector: flow-side operations; possible manifest sizes in
bytes: zero or one constant bytes;
storef - principal op code: store to memory; arithmetic domain: binary floating point;
addressing base: register; operation encoding group selector: flow-side operations;
possible manifest sizes in bytes: zero, one, two, or four constant bytes; operation
signature kind: base special register, manifest constant, belt operand from opsWindow,
scale factor for indexes in load/store/lea, late-evaluated operand from bel
storep - principal op code: store to memory; arithmetic domain: pointers; addressing base:
register; operation encoding group selector: flow-side operations; possible manifest
sizes in bytes: zero, one, two, or four constant bytes;
streamf - principal op code: allocate streamer frame; operation encoding group selector:
flow-side operations;
sub - Normal unsigned substraction.
sub1u - Decrement by one.
subd - Decimal floating point subtraction in current rounding mode.
subde - Decimal floating point subtraction in current rounding to even.
subdfz - Decimal floating point subtraction in current rounding away from zero.
subdn - Decimal floating point subtraction in current rounding to nearest.
subdp - Decimal floating point subtraction in current rounding towards positive infinity.
subdz - Decimal floating point subtraction in current rounding towards zero.
subf - Floating point subtraction in current rounding mode.
subfe - Floating point subtraction in current rounding to even.
subffz - Floating point subtraction in current rounding away from zero.
subfn - Floating point subtraction in current rounding to nearest.
subfp - Floating point subtraction in current rounding towards positive infinity.
subfz - Floating point subtraction in current rounding towards zero.
subp - Pointer subtraction. Protects the upper 4 reserved bits, can be scaled for indexing.
subs - Alias for sub.
subss - principal op code: subtraction; arithmetic domain: signed integer; overflow policy:
saturating; scalar vs. vector data: scalar; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
subssv - Alias for subss.
subsv - Alias for sub.
subsw - principal op code: subtraction; arithmetic domain: signed integer; overflow policy:
widening; scalar vs. vector data: scalar; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, small immediate constant of per-slot varying
rang
subswv - principal op code: subtraction; arithmetic domain: signed integer; overflow policy:
widening; scalar vs. vector data: vector; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
subsx - principal op code: subtraction; arithmetic domain: signed integer; overflow policy:
excepting; scalar vs. vector data: scalar; operation encoding group selector: exu-side
operations; condition code generator: generate condition codes; operation signature
kind: belt operand from ops window, belt operand from ops windo
subsxv - Alias for subsx.
subu - Alias for sub.
subus - principal op code: subtraction; arithmetic domain: unsigned integer; overflow
policy: saturating; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, small immediate constant of per-slot
varying rang
subusv - Alias for subus.
subuv - Alias for sub.
subuw - principal op code: subtraction; arithmetic domain: unsigned integer; overflow
policy: widening; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
subuwv - principal op code: subtraction; arithmetic domain: unsigned integer; overflow
policy: widening; scalar vs. vector data: vector; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, belt operand from ops windo
subux - principal op code: subtraction; arithmetic domain: unsigned integer; overflow
policy: excepting; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops window, small immediate constant of per-slot
varying rang
subuxv - Alias for subux.
test - Test if a single indexed bit is set.
u2fd - principal op code: convert unsigned integer to float; arithmetic domain: decimal
floating point; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
u2fde - principal op code: convert unsigned integer to float; arithmetic domain: decimal
floating point; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
u2fdfz - principal op code: convert unsigned integer to float; arithmetic domain: decimal
floating point; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
u2fdn - principal op code: convert unsigned integer to float; arithmetic domain: decimal
floating point; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
u2fdp - principal op code: convert unsigned integer to float; arithmetic domain: decimal
floating point; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
u2fdz - principal op code: convert unsigned integer to float; arithmetic domain: decimal
floating point; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops windo
u2ff - principal op code: convert unsigned integer to float; arithmetic domain: binary
floating point; rounding policy: use current dynamic rounding mode; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
u2ffe - principal op code: convert unsigned integer to float; arithmetic domain: binary
floating point; rounding policy: to nearest, ties toward even adjacent value; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
u2fffz - principal op code: convert unsigned integer to float; arithmetic domain: binary
floating point; rounding policy: to nearest, ties away from zero; operation encoding
group selector: exu-side operations; condition code generator: generate condition
codes;
u2ffn - principal op code: convert unsigned integer to float; arithmetic domain: binary
floating point; rounding policy: toward negative infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
u2ffp - principal op code: convert unsigned integer to float; arithmetic domain: binary
floating point; rounding policy: toward positive infinity; operation encoding group
selector: exu-side operations; condition code generator: generate condition codes;
u2ffz - principal op code: convert unsigned integer to float; arithmetic domain: binary
floating point; rounding policy: toward zero; operation encoding group selector:
exu-side operations; condition code generator: generate condition codes; operation
signature kind: belt operand from ops windo
u2s - principal op code: convert unsigned to signed; arithmetic domain: logical; overflow
policy: modulo; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations;
u2ss - principal op code: convert unsigned to signed; arithmetic domain: logical; overflow
policy: saturating; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations;
u2ssv - Alias for u2ss.
u2sv - Alias for u2s.
u2sw - principal op code: convert unsigned to signed; arithmetic domain: logical; overflow
policy: widening; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations;
u2swv - principal op code: convert unsigned to signed; arithmetic domain: logical; overflow
policy: widening; scalar vs. vector data: vector; operation encoding group selector:
exu-side operations;
u2sx - principal op code: convert unsigned to signed; arithmetic domain: logical; overflow
policy: excepting; scalar vs. vector data: scalar; operation encoding group selector:
exu-side operations;
u2sxv - Alias for u2sx.
vec - principal op code: vector constructor; arithmetic domain: logical; operation
encoding group selector: exu-side operations;
widen - principal op code: widen to double width; arithmetic domain: logical; scalar vs.
vector data: scalar; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops window, belt operand from ops windo
widend - principal op code: widen to double width; arithmetic domain: decimal floating point;
scalar vs. vector data: scalar; operation encoding group selector: exu-side operations;
widendv - principal op code: widen to double width; arithmetic domain: decimal floating point;
scalar vs. vector data: vector; operation encoding group selector: exu-side operations;
widenf - principal op code: widen to double width; arithmetic domain: binary floating point;
scalar vs. vector data: scalar; operation encoding group selector: exu-side operations;
widenfv - principal op code: widen to double width; arithmetic domain: binary floating point;
scalar vs. vector data: vector; operation encoding group selector: exu-side operations;
widens - principal op code: widen to double width; arithmetic domain: signed integer; scalar
vs. vector data: scalar; operation encoding group selector: exu-side operations;
widensf - principal op code: widen to double width; arithmetic domain: signed fixed point;
scalar vs. vector data: scalar; operation encoding group selector: exu-side operations;
widensfv - principal op code: widen to double width; arithmetic domain: signed fixed point;
scalar vs. vector data: vector; operation encoding group selector: exu-side operations;
widensv - principal op code: widen to double width; arithmetic domain: signed integer; scalar
vs. vector data: vector; operation encoding group selector: exu-side operations;
widenu - Alias for widen.
widenuf - principal op code: widen to double width; arithmetic domain: unsigned fixed point;
scalar vs. vector data: scalar; operation encoding group selector: exu-side operations;
widenufv - principal op code: widen to double width; arithmetic domain: unsigned fixed point;
scalar vs. vector data: vector; operation encoding group selector: exu-side operations;
widenuv - Alias for widenv.
widenv - principal op code: widen to double width; arithmetic domain: logical; scalar vs.
vector data: vector; operation encoding group selector: exu-side operations; operation
signature kind: belt operand from ops windo
wr - principal op code: hardware writer; arithmetic domain: logical; operation encoding
group selector: exu-side writers;
xorl - principal op code: bitwise exclusive or; arithmetic domain: logical; operation
encoding group selector: exu-side operations; condition code generator: generate
condition codes;
xorls - Alias for xorl.
xorlu - Alias for xorl.
