Signature is identical to fpmul:
v8float fpneg_mul(v32float xbuf, int xstart, unsigned int xoffs, v8float zbuf, int zstart, unsigned int zoffs)
It returns the opposite value of the product:
for (i = 0 ; i < 8 ; i++)
ret[i] = - xbuf[xstart + xoffs[i]] * zbuf[zstart + zoffs[i]]