@ -518,7 +518,8 @@ This function is a faster implementation of arm_mat_mult_q31 for MVE but it is r
</ul>
</ul>
</dd></dl>
</dd></dl>
<dlclass="section user"><dt>Scaling and Overflow Behavior</dt><dd>The function is implemented using an internal 64-bit accumulator. The inputs to the multiplications are in 1.15 format and multiplications yield a 2.30 result. The 2.30 intermediate results are accumulated in a 64-bit accumulator in 34.30 format. This approach provides 33 guard bits and there is no risk of overflow. The 34.30 result is then truncated to 34.15 format by discarding the low 15 bits and then saturated to 1.15 format. </dd></dl>
<dlclass="section user"><dt>Scaling and Overflow Behavior</dt><dd>The function is implemented using an internal 64-bit accumulator. The inputs to the multiplications are in 1.15 format and multiplications yield a 2.30 result. The 2.30 intermediate results are accumulated in a 64-bit accumulator in 34.30 format. This approach provides 33 guard bits and there is no risk of overflow. The 34.30 result is then truncated to 34.15 format by discarding the low 15 bits and then saturated to 1.15 format. </dd></dl>
<dlclass="section user"><dt></dt><dd>Refer to <aclass="el"href="group__MatrixMult.html#ga08f37d93a5bfef0c5000dc5e0a411f93">arm_mat_mult_fast_q15()</a> for a faster but less precise version of this function. </dd></dl>
<dlclass="section user"><dt></dt><dd>Refer to <aclass="el"href="group__MatrixMult.html#ga08f37d93a5bfef0c5000dc5e0a411f93">arm_mat_mult_fast_q15()</a> for a faster but less precise version of this function.</dd></dl>
<dlclass="section user"><dt>pState</dt><dd>pState will contain the transpose of pSrcB </dd></dl>
@ -795,46 +795,46 @@ Real Inverse Fast Fourier Transform</div></div>
<trclass="markdownTableHead">
<trclass="markdownTableHead">
<thclass="markdownTableHeadRight">RFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<thclass="markdownTableHeadRight">RFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<dlclass="section user"><dt>Input and Output formats for RIFFT Q15</dt><dd></dd></dl>
<dlclass="section user"><dt>Input and Output formats for RIFFT Q15</dt><dd></dd></dl>
<tableclass="markdownTable">
<tableclass="markdownTable">
<trclass="markdownTableHead">
<trclass="markdownTableHead">
<thclass="markdownTableHeadRight">RIFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<thclass="markdownTableHeadRight">RIFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<dlclass="section user"><dt></dt><dd>If the input buffer is of length N (fftLenReal), the output buffer must have length 2N since it is containing the conjugate part (except for MVE version where N+2 is enough). The input buffer is modified by this function. </dd></dl>
<dlclass="section user"><dt></dt><dd>If the input buffer is of length N (fftLenReal), the output buffer must have length 2N since it is containing the conjugate part (except for MVE version where N+2 is enough). The input buffer is modified by this function. </dd></dl>
<dlclass="section user"><dt></dt><dd>For the RIFFT, the source buffer must have length N+2 since the Nyquist frequency value is needed but conjugate part is ignored. It is not using the packing trick of the float version. </dd></dl>
<dlclass="section user"><dt></dt><dd>For the RIFFT, the source buffer must have length N+2 since the Nyquist frequency value is needed but conjugate part is ignored. It is not using the packing trick of the float version. </dd></dl>
@ -887,46 +887,46 @@ Real Inverse Fast Fourier Transform</div></div>
<trclass="markdownTableHead">
<trclass="markdownTableHead">
<thclass="markdownTableHeadRight">RFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<thclass="markdownTableHeadRight">RFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<dlclass="section user"><dt>Input and Output formats for RIFFT Q31</dt><dd></dd></dl>
<dlclass="section user"><dt>Input and Output formats for RIFFT Q31</dt><dd></dd></dl>
<tableclass="markdownTable">
<tableclass="markdownTable">
<trclass="markdownTableHead">
<trclass="markdownTableHead">
<thclass="markdownTableHeadRight">RIFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<thclass="markdownTableHeadRight">RIFFT Size </th><thclass="markdownTableHeadRight">Input Format </th><thclass="markdownTableHeadRight">Output Format </th><thclass="markdownTableHeadRight">Number of bits to upscale </th></tr>
<dlclass="section user"><dt></dt><dd>If the input buffer is of length N (fftLenReal), the output buffer must have length 2N since it is containing the conjugate part (except for MVE version where N+2 is enough). The input buffer is modified by this function. </dd></dl>
<dlclass="section user"><dt></dt><dd>If the input buffer is of length N (fftLenReal), the output buffer must have length 2N since it is containing the conjugate part (except for MVE version where N+2 is enough). The input buffer is modified by this function. </dd></dl>
<dlclass="section user"><dt></dt><dd>For the RIFFT, the source buffer must have length N+2 since the Nyquist frequency value is needed but conjugate part is ignored. It is not using the packing trick of the float version. </dd></dl>
<dlclass="section user"><dt></dt><dd>For the RIFFT, the source buffer must have length N+2 since the Nyquist frequency value is needed but conjugate part is ignored. It is not using the packing trick of the float version. </dd></dl>