Should we do `dred_compute_latents` in DTX?

We found that the average cost time doubled in silence segment compared with in voiced segment when enable DRED.

After checking the code, dred_compute_latents will always on even in DTX opus would send empty packet, and minor input will cause a huge increase in GRU's calculations.
To resolve this problem, we tried to enable OPUS_FLOAT_APPROX, but it didn't work, and then we add _mm_setcsr(csr | 0x8040) when do encode in our code, it works.

To significantly save our CPU, We want to ask you experts should we still do dred_compute_latents in DTX? To be specific, could we change to:

#ifdef ENABLE_DRED
    opus_int32 in_dtx = 0;
    if ( st->dred_duration > 0 && st->dred_encoder.loaded && opus_encoder_ctl(st, OPUS_GET_IN_DTX_REQUEST(in_dtx)) == OPUS_OK && !in_dtx ) {
        int frame_size_400Hz;
        /* DRED Encoder */
        dred_compute_latents( &st->dred_encoder, &pcm_buf[total_buffer*st->channels], frame_size, total_buffer, st->arch );
        frame_size_400Hz = frame_size*400/st->Fs;
        OPUS_MOVE(&st->activity_mem[frame_size_400Hz], st->activity_mem, 4*DRED_MAX_FRAMES-frame_size_400Hz);
        for (i=0;i<frame_size_400Hz;i++)
           st->activity_mem[i] = activity;
    } else {
        st->dred_encoder.latents_buffer_fill = 0;
        OPUS_CLEAR(st->activity_mem, DRED_MAX_FRAMES);
    }
#endif

???

Start by tracing the DRED encoder path around dred_compute_latents and the shown OPUS_GET_IN_DTX_REQUEST check. Reproduce the silence-versus-voiced CPU difference with DRED enabled, then verify DTX packet handling and latent/activity buffer behavior. Done means the project has a reviewed decision and tests or measurements covering the resulting DTX behavior.