AI Engine Intrinsics  (AIE) r2p21
 All Data Structures Namespaces Functions Variables Typedefs Groups Pages
Functions
Z buffer conjugation
32 bit Real x 16 bit Complex

Overview

32 bit real by 16 bit complex using Z buffer conjugation

Functions

v4cacc80 lmac4_abs_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc80 lmac4_abs_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmac4_max_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc80 lmac4_max_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmac4_max_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmac4_maxdiff_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc80 lmac4_maxdiff_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmac4_maxdiff_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmac4_min_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc80 lmac4_min_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmac4_min_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_abs_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc80 lmsc4_abs_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_max_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc80 lmsc4_max_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_max_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_maxdiff_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc80 lmsc4_maxdiff_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_maxdiff_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_min_c (v4cacc80 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc80 lmsc4_min_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmsc4_min_c (v4cacc80 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_abs_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc80 lmul4_abs_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_max_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc80 lmul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_maxdiff_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc80 lmul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_min_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc80 lmul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lmul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_abs_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc80 lnegmul4_abs_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_max_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc80 lnegmul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_maxdiff_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc80 lnegmul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_min_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc80 lnegmul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc80 lnegmul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_abs_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc48 mac4_abs_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_max_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc48 mac4_max_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_max_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_maxdiff_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc48 mac4_maxdiff_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_maxdiff_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_min_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc48 mac4_min_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mac4_min_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_abs_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v8cacc48 mac8_abs_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_max_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v8cacc48 mac8_max_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_max_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_maxdiff_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v8cacc48 mac8_maxdiff_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_maxdiff_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_min_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v8cacc48 mac8_min_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mac8_min_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_abs_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc48 msc4_abs_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_max_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc48 msc4_max_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_max_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_maxdiff_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc48 msc4_maxdiff_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_maxdiff_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_min_c (v4cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc48 msc4_min_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 msc4_min_c (v4cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_abs_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v8cacc48 msc8_abs_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_max_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v8cacc48 msc8_max_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_max_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_maxdiff_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v8cacc48 msc8_maxdiff_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_maxdiff_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_min_c (v8cacc48 acc, v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v8cacc48 msc8_min_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 msc8_min_c (v8cacc48 acc, v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_abs_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc48 mul4_abs_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_max_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc48 mul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_maxdiff_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc48 mul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_min_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc48 mul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 mul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_abs_c (v32int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v8cacc48 mul8_abs_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_max_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v8cacc48 mul8_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_maxdiff_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v8cacc48 mul8_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_min_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v8cacc48 mul8_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 mul8_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_abs_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v4cacc48 negmul4_abs_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_max_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v4cacc48 negmul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_maxdiff_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v4cacc48 negmul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_min_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v4cacc48 negmul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v4cacc48 negmul4_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int xstep, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets, int zstep)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_abs_c (v32int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .
 
v8cacc48 negmul8_abs_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_max_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .
 
v8cacc48 negmul8_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_max_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_maxdiff_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .
 
v8cacc48 negmul8_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_maxdiff_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_min_c (v32int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .
 
v8cacc48 negmul8_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 
v8cacc48 negmul8_min_c (v16int32 xbuff, int xstart, unsigned int xoffsets, v16int32 ybuff, int ystart, v8cint16 zbuff, int zstart, unsigned int zoffsets)
 Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.
 

Function Documentation

v4cacc80 lmac4_abs_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 += cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 += cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 += cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 += cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_abs_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 += cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 += cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 += cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_max_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 += cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 += cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 += cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 += cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_max_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 += cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 += cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 += cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_max_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 += cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 += cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 += cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_maxdiff_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_maxdiff_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_maxdiff_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_min_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 += cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 += cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 += cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 += cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_min_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 += cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 += cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 += cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmac4_min_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 += cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 += cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 += cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_abs_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 -= cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 -= cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 -= cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 -= cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_abs_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 -= cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 -= cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 -= cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_max_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 -= cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 -= cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 -= cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 -= cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_max_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 -= cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 -= cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 -= cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_max_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 -= cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 -= cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 -= cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_maxdiff_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_maxdiff_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_maxdiff_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_min_c ( v4cacc80  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 -= cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 -= cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 -= cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 -= cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_min_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 -= cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 -= cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 -= cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmsc4_min_c ( v4cacc80  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 -= cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 -= cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 -= cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
acc v4cacc80 Incoming accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_abs_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 = cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 = cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 = cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 = cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_abs_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 = cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 = cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 = cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_max_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 = cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 = cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 = cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 = cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 = cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 = cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 = cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 = cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 = cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 = cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_maxdiff_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_min_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 = cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 = cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 = cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 = cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 = cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 = cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 = cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lmul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 = cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 = cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 = cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_abs_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 = -( cnj(z00)*abs(x00) + cnj(z01)*abs(x01) )
acc1 = -( cnj(z10)*abs(x10) + cnj(z11)*abs(x11) )
acc2 = -( cnj(z20)*abs(x20) + cnj(z21)*abs(x21) )
acc3 = -( cnj(z30)*abs(x30) + cnj(z31)*abs(x31) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_abs_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*abs(x00) + cnj(z01)*abs(x01) )
acc1 = -( cnj(z10)*abs(x10) + cnj(z11)*abs(x11) )
acc2 = -( cnj(z20)*abs(x20) + cnj(z21)*abs(x21) )
acc3 = -( cnj(z30)*abs(x30) + cnj(z31)*abs(x31) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_max_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_maxdiff_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_min_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc80 lnegmul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc80 Returned accumulation vector (4 x cint80 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_abs_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 += cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 += cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 += cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 += cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_abs_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 += cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 += cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 += cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_max_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 += cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 += cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 += cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 += cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_max_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 += cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 += cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 += cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_max_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 += cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 += cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 += cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_maxdiff_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_maxdiff_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_maxdiff_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_min_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 += cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 += cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 += cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 += cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_min_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 += cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 += cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 += cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mac4_min_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 += cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 += cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 += cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_abs_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 += cnj(z00)*abs(x00)
acc1 += cnj(z10)*abs(x10)
acc2 += cnj(z20)*abs(x20)
acc3 += cnj(z30)*abs(x30)
acc4 += cnj(z40)*abs(x40)
acc5 += cnj(z50)*abs(x50)
acc6 += cnj(z60)*abs(x60)
acc7 += cnj(z70)*abs(x70)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_abs_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*abs(x00)
acc1 += cnj(z10)*abs(x10)
acc2 += cnj(z20)*abs(x20)
acc3 += cnj(z30)*abs(x30)
acc4 += cnj(z40)*abs(x40)
acc5 += cnj(z50)*abs(x50)
acc6 += cnj(z60)*abs(x60)
acc7 += cnj(z70)*abs(x70)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_max_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 += cnj(z00)*(max(x00 , y00))
acc1 += cnj(z10)*(max(x10 , y10))
acc2 += cnj(z20)*(max(x20 , y20))
acc3 += cnj(z30)*(max(x30 , y30))
acc4 += cnj(z40)*(max(x40 , y40))
acc5 += cnj(z50)*(max(x50 , y50))
acc6 += cnj(z60)*(max(x60 , y60))
acc7 += cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_max_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(max(x00 , y00))
acc1 += cnj(z10)*(max(x10 , y10))
acc2 += cnj(z20)*(max(x20 , y20))
acc3 += cnj(z30)*(max(x30 , y30))
acc4 += cnj(z40)*(max(x40 , y40))
acc5 += cnj(z50)*(max(x50 , y50))
acc6 += cnj(z60)*(max(x60 , y60))
acc7 += cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_max_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(max(x00 , y00))
acc1 += cnj(z10)*(max(x10 , y10))
acc2 += cnj(z20)*(max(x20 , y20))
acc3 += cnj(z30)*(max(x30 , y30))
acc4 += cnj(z40)*(max(x40 , y40))
acc5 += cnj(z50)*(max(x50 , y50))
acc6 += cnj(z60)*(max(x60 , y60))
acc7 += cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_maxdiff_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 += cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 += cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 += cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 += cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_maxdiff_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 += cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 += cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 += cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 += cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_maxdiff_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 += cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 += cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 += cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 += cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 += cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 += cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 += cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_min_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 += cnj(z00)*(min(x00 , y00))
acc1 += cnj(z10)*(min(x10 , y10))
acc2 += cnj(z20)*(min(x20 , y20))
acc3 += cnj(z30)*(min(x30 , y30))
acc4 += cnj(z40)*(min(x40 , y40))
acc5 += cnj(z50)*(min(x50 , y50))
acc6 += cnj(z60)*(min(x60 , y60))
acc7 += cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_min_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(min(x00 , y00))
acc1 += cnj(z10)*(min(x10 , y10))
acc2 += cnj(z20)*(min(x20 , y20))
acc3 += cnj(z30)*(min(x30 , y30))
acc4 += cnj(z40)*(min(x40 , y40))
acc5 += cnj(z50)*(min(x50 , y50))
acc6 += cnj(z60)*(min(x60 , y60))
acc7 += cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mac8_min_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-accumulate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 += cnj(z00)*(min(x00 , y00))
acc1 += cnj(z10)*(min(x10 , y10))
acc2 += cnj(z20)*(min(x20 , y20))
acc3 += cnj(z30)*(min(x30 , y30))
acc4 += cnj(z40)*(min(x40 , y40))
acc5 += cnj(z50)*(min(x50 , y50))
acc6 += cnj(z60)*(min(x60 , y60))
acc7 += cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_abs_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 -= cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 -= cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 -= cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 -= cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_abs_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 -= cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 -= cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 -= cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_max_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 -= cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 -= cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 -= cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 -= cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_max_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 -= cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 -= cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 -= cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_max_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 -= cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 -= cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 -= cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_maxdiff_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_maxdiff_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_maxdiff_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_min_c ( v4cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 -= cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 -= cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 -= cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 -= cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_min_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 -= cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 -= cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 -= cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 msc4_min_c ( v4cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 -= cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 -= cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 -= cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
acc v4cacc48 Incoming accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_abs_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 -= cnj(z00)*abs(x00)
acc1 -= cnj(z10)*abs(x10)
acc2 -= cnj(z20)*abs(x20)
acc3 -= cnj(z30)*abs(x30)
acc4 -= cnj(z40)*abs(x40)
acc5 -= cnj(z50)*abs(x50)
acc6 -= cnj(z60)*abs(x60)
acc7 -= cnj(z70)*abs(x70)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_abs_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*abs(x00)
acc1 -= cnj(z10)*abs(x10)
acc2 -= cnj(z20)*abs(x20)
acc3 -= cnj(z30)*abs(x30)
acc4 -= cnj(z40)*abs(x40)
acc5 -= cnj(z50)*abs(x50)
acc6 -= cnj(z60)*abs(x60)
acc7 -= cnj(z70)*abs(x70)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_max_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 -= cnj(z00)*(max(x00 , y00))
acc1 -= cnj(z10)*(max(x10 , y10))
acc2 -= cnj(z20)*(max(x20 , y20))
acc3 -= cnj(z30)*(max(x30 , y30))
acc4 -= cnj(z40)*(max(x40 , y40))
acc5 -= cnj(z50)*(max(x50 , y50))
acc6 -= cnj(z60)*(max(x60 , y60))
acc7 -= cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_max_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(max(x00 , y00))
acc1 -= cnj(z10)*(max(x10 , y10))
acc2 -= cnj(z20)*(max(x20 , y20))
acc3 -= cnj(z30)*(max(x30 , y30))
acc4 -= cnj(z40)*(max(x40 , y40))
acc5 -= cnj(z50)*(max(x50 , y50))
acc6 -= cnj(z60)*(max(x60 , y60))
acc7 -= cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_max_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(max(x00 , y00))
acc1 -= cnj(z10)*(max(x10 , y10))
acc2 -= cnj(z20)*(max(x20 , y20))
acc3 -= cnj(z30)*(max(x30 , y30))
acc4 -= cnj(z40)*(max(x40 , y40))
acc5 -= cnj(z50)*(max(x50 , y50))
acc6 -= cnj(z60)*(max(x60 , y60))
acc7 -= cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_maxdiff_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 -= cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 -= cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 -= cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 -= cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_maxdiff_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 -= cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 -= cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 -= cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 -= cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_maxdiff_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 -= cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 -= cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 -= cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 -= cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 -= cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 -= cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 -= cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_min_c ( v8cacc48  acc,
v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 -= cnj(z00)*(min(x00 , y00))
acc1 -= cnj(z10)*(min(x10 , y10))
acc2 -= cnj(z20)*(min(x20 , y20))
acc3 -= cnj(z30)*(min(x30 , y30))
acc4 -= cnj(z40)*(min(x40 , y40))
acc5 -= cnj(z50)*(min(x50 , y50))
acc6 -= cnj(z60)*(min(x60 , y60))
acc7 -= cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_min_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(min(x00 , y00))
acc1 -= cnj(z10)*(min(x10 , y10))
acc2 -= cnj(z20)*(min(x20 , y20))
acc3 -= cnj(z30)*(min(x30 , y30))
acc4 -= cnj(z40)*(min(x40 , y40))
acc5 -= cnj(z50)*(min(x50 , y50))
acc6 -= cnj(z60)*(min(x60 , y60))
acc7 -= cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 msc8_min_c ( v8cacc48  acc,
v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-subtract intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 -= cnj(z00)*(min(x00 , y00))
acc1 -= cnj(z10)*(min(x10 , y10))
acc2 -= cnj(z20)*(min(x20 , y20))
acc3 -= cnj(z30)*(min(x30 , y30))
acc4 -= cnj(z40)*(min(x40 , y40))
acc5 -= cnj(z50)*(min(x50 , y50))
acc6 -= cnj(z60)*(min(x60 , y60))
acc7 -= cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
acc v8cacc48 Incoming accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_abs_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 = cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 = cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 = cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 = cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_abs_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*abs(x00) + cnj(z01)*abs(x01)
acc1 = cnj(z10)*abs(x10) + cnj(z11)*abs(x11)
acc2 = cnj(z20)*abs(x20) + cnj(z21)*abs(x21)
acc3 = cnj(z30)*abs(x30) + cnj(z31)*abs(x31)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_max_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 = cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 = cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 = cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 = cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 = cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 = cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 = cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01))
acc1 = cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11))
acc2 = cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21))
acc3 = cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_maxdiff_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0)

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_min_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 = cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 = cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 = cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 = cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 = cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 = cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 = cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 mul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01))
acc1 = cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11))
acc2 = cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21))
acc3 = cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31))

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_abs_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 = cnj(z00)*abs(x00)
acc1 = cnj(z10)*abs(x10)
acc2 = cnj(z20)*abs(x20)
acc3 = cnj(z30)*abs(x30)
acc4 = cnj(z40)*abs(x40)
acc5 = cnj(z50)*abs(x50)
acc6 = cnj(z60)*abs(x60)
acc7 = cnj(z70)*abs(x70)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_abs_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*abs(x00)
acc1 = cnj(z10)*abs(x10)
acc2 = cnj(z20)*abs(x20)
acc3 = cnj(z30)*abs(x30)
acc4 = cnj(z40)*abs(x40)
acc5 = cnj(z50)*abs(x50)
acc6 = cnj(z60)*abs(x60)
acc7 = cnj(z70)*abs(x70)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_max_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 = cnj(z00)*(max(x00 , y00))
acc1 = cnj(z10)*(max(x10 , y10))
acc2 = cnj(z20)*(max(x20 , y20))
acc3 = cnj(z30)*(max(x30 , y30))
acc4 = cnj(z40)*(max(x40 , y40))
acc5 = cnj(z50)*(max(x50 , y50))
acc6 = cnj(z60)*(max(x60 , y60))
acc7 = cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(max(x00 , y00))
acc1 = cnj(z10)*(max(x10 , y10))
acc2 = cnj(z20)*(max(x20 , y20))
acc3 = cnj(z30)*(max(x30 , y30))
acc4 = cnj(z40)*(max(x40 , y40))
acc5 = cnj(z50)*(max(x50 , y50))
acc6 = cnj(z60)*(max(x60 , y60))
acc7 = cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(max(x00 , y00))
acc1 = cnj(z10)*(max(x10 , y10))
acc2 = cnj(z20)*(max(x20 , y20))
acc3 = cnj(z30)*(max(x30 , y30))
acc4 = cnj(z40)*(max(x40 , y40))
acc5 = cnj(z50)*(max(x50 , y50))
acc6 = cnj(z60)*(max(x60 , y60))
acc7 = cnj(z70)*(max(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_maxdiff_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 = cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 = cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 = cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 = cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 = cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 = cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 = cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 = cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0)
acc1 = cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0)
acc2 = cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0)
acc3 = cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0)
acc4 = cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0)
acc5 = cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0)
acc6 = cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0)
acc7 = cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0)

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_min_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 = cnj(z00)*(min(x00 , y00))
acc1 = cnj(z10)*(min(x10 , y10))
acc2 = cnj(z20)*(min(x20 , y20))
acc3 = cnj(z30)*(min(x30 , y30))
acc4 = cnj(z40)*(min(x40 , y40))
acc5 = cnj(z50)*(min(x50 , y50))
acc6 = cnj(z60)*(min(x60 , y60))
acc7 = cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(min(x00 , y00))
acc1 = cnj(z10)*(min(x10 , y10))
acc2 = cnj(z20)*(min(x20 , y20))
acc3 = cnj(z30)*(min(x30 , y30))
acc4 = cnj(z40)*(min(x40 , y40))
acc5 = cnj(z50)*(min(x50 , y50))
acc6 = cnj(z60)*(min(x60 , y60))
acc7 = cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 mul8_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = cnj(z00)*(min(x00 , y00))
acc1 = cnj(z10)*(min(x10 , y10))
acc2 = cnj(z20)*(min(x20 , y20))
acc3 = cnj(z30)*(min(x30 , y30))
acc4 = cnj(z40)*(min(x40 , y40))
acc5 = cnj(z50)*(min(x50 , y50))
acc6 = cnj(z60)*(min(x60 , y60))
acc7 = cnj(z70)*(min(x70 , y70))

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_abs_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 = -( cnj(z00)*abs(x00) + cnj(z01)*abs(x01) )
acc1 = -( cnj(z10)*abs(x10) + cnj(z11)*abs(x11) )
acc2 = -( cnj(z20)*abs(x20) + cnj(z21)*abs(x21) )
acc3 = -( cnj(z30)*abs(x30) + cnj(z31)*abs(x31) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_abs_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*abs(x00) + cnj(z01)*abs(x01) )
acc1 = -( cnj(z10)*abs(x10) + cnj(z11)*abs(x11) )
acc2 = -( cnj(z20)*abs(x20) + cnj(z21)*abs(x21) )
acc3 = -( cnj(z30)*abs(x30) + cnj(z31)*abs(x31) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
xstep unsigned int Step between each column for selection in the xbuffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_max_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(max(x00 , y00)) + cnj(z01)*(max(x01 , y01)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) + cnj(z11)*(max(x11 , y11)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) + cnj(z21)*(max(x21 , y21)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) + cnj(z31)*(max(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_maxdiff_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) + cnj(z01)*((x01 - y01 > 0) ? (x01 - y01):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) + cnj(z11)*((x11 - y11 > 0) ? (x11 - y11):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) + cnj(z21)*((x21 - y21 > 0) ? (x21 - y21):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) + cnj(z31)*((x31 - y31 > 0) ? (x31 - y31):0) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_min_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v4cacc48 negmul4_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  xstep,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets,
int  zstep 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(min(x00 , y00)) + cnj(z01)*(min(x01 , y01)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) + cnj(z11)*(min(x11 , y11)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) + cnj(z21)*(min(x21 , y21)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) + cnj(z31)*(min(x31 , y31)) )

Parameters

Input/OutputType Comments
return v4cacc48 Returned accumulation vector (4 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
xstep unsigned int Step between each column for selection in the x and y buffers. Ystep is symmetric to xstep (ystep advances by -xstep).
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsetsunsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
zstep int Step between each column for selection in the zbuffer.
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_abs_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer .

acc0 = -( cnj(z00)*abs(x00) )
acc1 = -( cnj(z10)*abs(x10) )
acc2 = -( cnj(z20)*abs(x20) )
acc3 = -( cnj(z30)*abs(x30) )
acc4 = -( cnj(z40)*abs(x40) )
acc5 = -( cnj(z50)*abs(x50) )
acc6 = -( cnj(z60)*abs(x60) )
acc7 = -( cnj(z70)*abs(x70) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_abs_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the absolute value in the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*abs(x00) )
acc1 = -( cnj(z10)*abs(x10) )
acc2 = -( cnj(z20)*abs(x20) )
acc3 = -( cnj(z30)*abs(x30) )
acc4 = -( cnj(z40)*abs(x40) )
acc5 = -( cnj(z50)*abs(x50) )
acc6 = -( cnj(z60)*abs(x60) )
acc7 = -( cnj(z70)*abs(x70) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to the x buffer. LSB apply to first lane
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_max_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*(max(x00 , y00)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) )
acc4 = -( cnj(z40)*(max(x40 , y40)) )
acc5 = -( cnj(z50)*(max(x50 , y50)) )
acc6 = -( cnj(z60)*(max(x60 , y60)) )
acc7 = -( cnj(z70)*(max(x70 , y70)) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(max(x00 , y00)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) )
acc4 = -( cnj(z40)*(max(x40 , y40)) )
acc5 = -( cnj(z50)*(max(x50 , y50)) )
acc6 = -( cnj(z60)*(max(x60 , y60)) )
acc7 = -( cnj(z70)*(max(x70 , y70)) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_max_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(max(x00 , y00)) )
acc1 = -( cnj(z10)*(max(x10 , y10)) )
acc2 = -( cnj(z20)*(max(x20 , y20)) )
acc3 = -( cnj(z30)*(max(x30 , y30)) )
acc4 = -( cnj(z40)*(max(x40 , y40)) )
acc5 = -( cnj(z50)*(max(x50 , y50)) )
acc6 = -( cnj(z60)*(max(x60 , y60)) )
acc7 = -( cnj(z70)*(max(x70 , y70)) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_maxdiff_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) )
acc4 = -( cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0) )
acc5 = -( cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0) )
acc6 = -( cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0) )
acc7 = -( cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) )
acc4 = -( cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0) )
acc5 = -( cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0) )
acc6 = -( cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0) )
acc7 = -( cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_maxdiff_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the maximum difference between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*((x00 - y00 > 0) ? (x00 - y00):0) )
acc1 = -( cnj(z10)*((x10 - y10 > 0) ? (x10 - y10):0) )
acc2 = -( cnj(z20)*((x20 - y20 > 0) ? (x20 - y20):0) )
acc3 = -( cnj(z30)*((x30 - y30 > 0) ? (x30 - y30):0) )
acc4 = -( cnj(z40)*((x40 - y40 > 0) ? (x40 - y40):0) )
acc5 = -( cnj(z50)*((x50 - y50 > 0) ? (x50 - y50):0) )
acc6 = -( cnj(z60)*((x60 - y60 > 0) ? (x60 - y60):0) )
acc7 = -( cnj(z70)*((x70 - y70 > 0) ? (x70 - y70):0) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_min_c ( v32int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer .

acc0 = -( cnj(z00)*(min(x00 , y00)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) )
acc4 = -( cnj(z40)*(min(x40 , y40)) )
acc5 = -( cnj(z50)*(min(x50 , y50)) )
acc6 = -( cnj(z60)*(min(x60 , y60)) )
acc7 = -( cnj(z70)*(min(x70 , y70)) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v32int32Input buffer of 32 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(min(x00 , y00)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) )
acc4 = -( cnj(z40)*(min(x40 , y40)) )
acc5 = -( cnj(z50)*(min(x50 , y50)) )
acc6 = -( cnj(z60)*(min(x60 , y60)) )
acc7 = -( cnj(z70)*(min(x70 , y70)) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.
v8cacc48 negmul8_min_c ( v16int32  xbuff,
int  xstart,
unsigned int  xoffsets,
v16int32  ybuff,
int  ystart,
v8cint16  zbuff,
int  zstart,
unsigned int  zoffsets 
)

Complex multiply-negate intrinsic function with z buffer conjugation after computing the minimum between the selected lanes from the input buffer using small X input buffer.

acc0 = -( cnj(z00)*(min(x00 , y00)) )
acc1 = -( cnj(z10)*(min(x10 , y10)) )
acc2 = -( cnj(z20)*(min(x20 , y20)) )
acc3 = -( cnj(z30)*(min(x30 , y30)) )
acc4 = -( cnj(z40)*(min(x40 , y40)) )
acc5 = -( cnj(z50)*(min(x50 , y50)) )
acc6 = -( cnj(z60)*(min(x60 , y60)) )
acc7 = -( cnj(z70)*(min(x70 , y70)) )

Parameters

Input/OutputType Comments
return v8cacc48 Returned accumulation vector (8 x cint48 lanes)
xbuff v16int32Input buffer of 16 elements of type int32
xstart int Starting position offset applied to all lanes of input from X buffer.
xoffsets unsigned int 4b offset for each lane, applied to both x and y buffers. LSB apply to first lane
ybuff v16int32Right input buffer of 16 elements of type int32
ystart int Starting position offset applied to all lanes for input from Y buffer.
zbuff v8cint16Input buffer of 8 elements of type cint16
zstart int Starting position offset applied to all lanes for input from Z buffer. This must be a compile time constant. Only the 4 LSB of the argument are used.
zoffsets unsigned int 4b offset for each lane, applied to input from Z buffer. LSB apply to first lane
Note
  • For more information on how data selection works from the buffers go here. The data buffer and coefficient buffer in this intrinsic use the general scheme.
  • Parameter 'zstart' must be a compile time constant.