Removes `experimental` from `experimental_fixed_format_kernels` flag

Renames `experimental_fixed_format_kernels` build option to
`fixed_format_kernels`.

Adds documentation for the flag covering basics:
- What fixed-format kernels are
- Why they're needed
- Which backend they're for (i.e. CPU)
- Some pointers on how to use them.

Resolves: ONCPUML-1253
Change-Id: I428c98614c309c9ffc32d0f32daa24740f7cb967
Signed-off-by: Nathan John Sircombe <nathan.sircombe@arm.com>
Reviewed-on: https://review.mlplatform.org/c/ml/ComputeLibrary/+/9523
Benchmark: Arm Jenkins <bsgcomp@arm.com>
Tested-by: Arm Jenkins <bsgcomp@arm.com>
Reviewed-by: SiCong Li <sicong.li@arm.com>
Comments-Addressed: Arm Jenkins <bsgcomp@arm.com>

1 files changed, 14 insertions, 0 deletions

diff --git a/docs/user_guide/how_to_build_and_run_examples.dox b/docs/user_guide/how_to_build_and_run_examples.dox
index 8aab445093..e0079cf42a 100644
--- a/docs/user_guide/how_to_build_and_run_examples.dox
+++ b/docs/user_guide/how_to_build_and_run_examples.dox
@@ -513,5 +513,19 @@ To build libraries, examples and tests:
 	cmake .. -DOPENMP=1 -DWERROR=0 -DDEBUG=0 -DBUILD_EXAMPLES=1 -DBUILD_TESTING=1 -DCMAKE_INSTALL_LIBDIR=.
 	cmake --build . -j32
 
+@section S1_8_fixed_format Building with support for fixed format kernels
+
+@subsection S1_8_1_intro_to_fixed_format_kernels What are fixed format kernels?
+
+The GEMM kernels used for convolutions and fully-connected layers in Compute Library employ memory layouts optimized for each kernel implementation. This then requires the supplied weights to be re-ordered into a buffer ready for consumption by the GEMM kernel. Where Compute Library is being called from a framework or library which implements operator caching, the re-ordering of the inputted weights into an intermediate buffer may no longer be desirable. When using a cached operator, the caller may wish to re-write the weights tensor, and re-run the operator using the updated weights. With the default GEMM kernels in Compute Library, the GEMM will be executed with the old weights, leading to incorrect results.
+
+To address this, Compute Library provides a set of GEMM kernels which use a common blocked memory format. These kernels consume the input weights directly from the weights buffer and do not execute an intermediate pre-transpose step. With this approach, it is the responsibility of the user (in this case the calling framework) to ensure that the weights are re-ordered into the required memory format. @ref NEGEMM::has_opt_impl is a static function that queries whether there exists fixed-format kernel, and if so will return in the expected weights format. The supported weight formats are enumerated in @ref arm_compute::WeightFormat.
+
+@subsection S1_8_2_building_fixed_format Building with fixed format kernels
+
+Fixed format kernels are only available for the CPU backend. To build Compute Library with fixed format kernels set fixed_format_kernels=1:
+
+        scons Werror=1 debug=0 neon=1 opencl=0 embed_kernels=0 os=linux multi_isa=1 build=native cppthreads=1 openmp=0 fixed_format_kernels=1
+
 */
 } // namespace arm_compute