Jean-Baptiste Rolland
3bf0f6e814
Enable f64 Neon optimisations only for aarch64
3 years ago
Jean-Baptiste Rolland
6bfb11f70d
Replace tabs with space
3 years ago
Silfurion
433ecc8b22
Updated the indentation and added the optimized var function
3 years ago
Silfurion
e5753198af
Added functions optimised for 64 bit float
3 years ago
Christophe Favergeon
5721f94b28
Improved documentation
...
Corrected issues #7 and #8
Removed duplicates
Started using Mathjax
3 years ago
Christophe Favergeon
3f95fbca89
New repository for CMSIS-DSP
4 years ago
Johan Bengtsson
065e8351c3
DSP: Remove undefined behavior in post-increment/decrement read ( #1443 )
...
According to the C standard q15_t * and const q15_t * are not compatible
types which, among other things, imply that an object of type const
q15_t * can't be modified by writing to a q15_t ** or vice versa.
Programs doing this are undefined.
Because of this rule all programs using the functions read_q15x2_ia,
read_q15x2_da, read_q7x4_ia, or read_q7x4_da for reading data from an
array of constant elements will be undefined. To solve this it is not
enough to change the type of the function since this will give problems
when reading data from an array of non-const elements. To get a defined
solution I needed to switch from functions to macros to allow the
increment to be done in the original type of the pointer.
4 years ago
JbR
d25b745082
Fix warning unused tmp variable ( #1335 )
...
* Fix warning unused tmp variable
* Fix comments
4 years ago
JbR
329dae5fad
Removing defgroup from f64 & fixing double init values
4 years ago
JbR
4954674848
Fixing double init value
4 years ago
JbR
7a4579d9a9
Implement some f64 methods
4 years ago
Christophe Favergeon
079b404a19
CMSIS-DSP : Correcting issue 1248
...
ARM_MATH_DSP is not behind a define guard
4 years ago
Christophe Favergeon
5fac45cc96
CMSIS-DSP: Removed float promotion issues
...
Corrected some additional warnings when compiling with AC5 compiler.
Updated version numbers,
5 years ago
Christophe Favergeon
d5ad8e995b
CMSIS-DSP: Update copyright year and version numbers in comments.
5 years ago
Christophe Favergeon
68dd797c99
CMSIS-DSP: Helium code for clipping functions.
5 years ago
Christophe Favergeon
0556e5fb2a
CMSIS-DSP: Added scalar version of clipping functions.
5 years ago
Christophe Favergeon
9a254bc926
CMSIS-DSP: Improvements for building with gcc on M55.
...
Solve most of f16 issues. But there are still some remaining
build issues with gcc10q4.
2 functions are reverting to scalar version when build with gcc on M55.
(Since Helium versions of those functions are not building).
5 years ago
killua-killua
f657d52568
Update arm_shift_q15.c
...
type promotion before __SSAT
5 years ago
killua-killua
d18f717c68
Update arm_shift_q7.c
...
type promotion before __SSAT
5 years ago
killua-killua
dc096e0de2
Update arm_offset_q31.c
...
It seems that there is __QADD instruction in all cases. Can we just use __QADD here, like you did in arm_add_q31.c .
5 years ago
killua-killua
0dc176154c
Update arm_offset_q7.c
...
type promotion before __SSAT
5 years ago
Yin-Cheng-888
1ca28fa995
CMSIS-DSP: Fix signed/unsigned integer data type issue
5 years ago
Christophe Favergeon
0bcb1384f2
CMSIS-DSP: More consistent use of the flag ARM_MATH_AUTOVECTORIZE
5 years ago
Christophe Favergeon
68e5b742ba
CMSIS-DSP: Some improvement to a few f16 functions
5 years ago
Christophe Favergeon
41f2015387
CMSIS-DSP: Re-organization of arm_math.h
...
arm_math.h splitted into several headers.
Interpolation functions moved from arm_math.h to a separate folder.
5 years ago
Christophe Favergeon
c0d3b61f37
CMSIS-DSP: Changed cmake required minimum version
6 years ago
Christophe Favergeon
be40746c5e
CMSIS-DSP: Improvement to test framework
...
Added support for A32.
Still some issues with cycle model.
6 years ago
Felix Johnny
23d216b3bb
CMSIS-NN: Fix warnings
...
1. Doxygen warnings are fixed
2. Sign conversion warnings(-Wsign-conversion) and -Wextra warnings
are fixed for TFLu compliant MVE APIs
Change-Id: I22ee8cd4ff4c94c92d9659504fc4c24d2b5ae441
6 years ago
Christophe Favergeon
a159791720
CMSIS-DSP: Improved test framework
...
Added gcc support to IPSS in test framework.
Improved f16 handling for when compiling M55 with gcc.
6 years ago
Christophe Favergeon
c33adc9485
CMSIS-DSP: Improvements to the float16 support for building.
6 years ago
Christophe Favergeon
b543f5c61c
CMSIS-DSP: Added f16 support to test framework.
...
Added preliminary f16 versions of some functions
6 years ago
Christophe Favergeon
017962caec
CMSIS-DSP: Fix for issue 874
6 years ago
Jonatan Antoni
9ec9907e1d
CMSIS-DSP: Updates pack source component.
...
- Bumped component version to 1.8.0
- Added new groups/functions/files
Change-Id: Iee15d6af1e7e9c1169867270ca04f9ca7d5552c7
6 years ago
ClaudioMartino
64d43685fb
CMSIS-DSP: Added MVE support for logical operators (modified prototypes)
6 years ago
Christophe Favergeon
c5339c7910
CMSIS-DSP: Corrected Doxygen issues
...
Issues with documentation of elementwise functions;
Issues with sorting functions;
Issues with splines functions.
6 years ago
Christophe Favergeon
d2fb32de54
CMSIS-DSP: Added longer test patterns and corrected saturation bugs.
...
Longer test patterns for BasicMaths, ComplexMaths and Statistics.
Corrected saturation bug in MVE code for arm_abs_q7,
arm_cmplx_conj_q15 and arm_cmplx_conj_q31.
6 years ago
Christophe Favergeon
4e4b9ca8bb
CMSIS-DSP: Corrected build issues with AC5 and gcc. Improved Doxygen comments.
6 years ago
ClaudioMartino
bbafcb8960
Added logical operators + sorting + spline
...
- Added NEON bitwise AND, NOT, OR, XOR (q7, q15, q31)
- Added Sorting algorithms f32 (NEON bitonic sort)
- Added cubic spline interpolation function
- Added test patterns for all
6 years ago
ClaudioMartino
07a6ef535d
Added arm_abs_q31.c Neon
6 years ago
Christophe Favergeon
c19c41d261
CMSIS-DSP: Added a ARM_MATH_FLOAT16 compilation option.
...
It is automatically enabled when MVE support is enabled.
6 years ago
Christophe Favergeon
2e12da1043
CMSIS-DSP: Corrected compilation warnings
...
More compilation tests done with AC5
6 years ago
Christophe Favergeon
8526741f9e
CMSIS-DSP: Added Helium versions of Fast Math, Support, SVM, Naive Bayes, Distance and additional Statistics functions.
...
Added a new header of vectorial inline functions for Neon and Helium.
This header is used in the implementation of some functions.
Corrected used of float (float functions rather than double ones).
6 years ago
Christophe Favergeon
56a540336a
CMSIS-DSP: Added Helium support for functions in BasicMaths.
...
Added test patterns for functions in BasicMaths.
Corrected github issue 622 (clarified comment for lms_norm_f32).
Added support for q63_t to the test framework.
6 years ago
Christophe Favergeon
84c7261036
CMSIS-DSP: Simplification to cmake to build without test framework.
6 years ago
Christophe Favergeon
4a0aa2ad26
CMSIS-DSP: Added support for Helium.
...
Only arm_dot_prod_f32 is currently providing an Helium implementation.
6 years ago
Christophe Favergeon
8a4c63e4d8
CMSIS-DSP: New testing framework
...
(For our internal use. In short term, we won't give support about it).
CMSIS-DSP: Update to cmake build for the testing framework
CMSIS-NN:Implementation of arm_fully_connected_s8
Use API and quantization compatible with TF Lite.
7 years ago
ClaudioMartino
ad726aa823
vmlaq in arm_dot_prod_f32 instead of vmulq and vadd
7 years ago
Christophe Favergeon
080b994dd7
CMSIS-DSP: Customization of FFT tables.
...
Compilation symbols have been introduced to allow to include only the
FFT tables which are useful in a given application.
The README.md in DSP is giving more details.
7 years ago
ClaudioMartino
7c1fca8052
Removed comments about loop unrolling in NEON codes to avoid confusion
7 years ago
ClaudioMartino
5792f80cf1
New NEON float32 algorithms for Cortex A5 (TransformFunctions excluded) with armDS project
7 years ago