diff options
author | Tai Ly <tai.ly@arm.com> | 2023-03-28 22:06:56 +0000 |
---|---|---|
committer | Tai Ly <tai.ly@arm.com> | 2023-05-05 19:23:15 +0000 |
commit | a4d748b08accce06fab93e2d2b96e499b35ae89b (patch) | |
tree | 20a3957e1f45f65f35d5d67ecce1618659e388f0 /reference_model/include/func_config.h | |
parent | 0c71686875618b2e11290273b7a05b88ef8a8aae (diff) | |
download | reference_model-a4d748b08accce06fab93e2d2b96e499b35ae89b.tar.gz |
[reference model] Add precise mode
This adds --precise_mode=1 option to tosa_referece_model,
which will cause reference model to convert all floating point tensors
to FP64 tensors and compute all operators accordingly.
Also adds optional -p arguments to test runners tosa_verif_run_tests.py
and tosa_verif_framework_compiler_runner.py to run tests in precise mode
Signed-off-by: Tai Ly <tai.ly@arm.com>
Change-Id: I156055216ad61710096497a8fa1a653be2a602a3
Diffstat (limited to 'reference_model/include/func_config.h')
-rw-r--r-- | reference_model/include/func_config.h | 1 |
1 files changed, 1 insertions, 0 deletions
diff --git a/reference_model/include/func_config.h b/reference_model/include/func_config.h index c1f8ef6..b92845b 100644 --- a/reference_model/include/func_config.h +++ b/reference_model/include/func_config.h @@ -48,6 +48,7 @@ struct func_config_t uint32_t tosa_profile = 1; uint32_t dump_intermediates = 0; std::string fp_format = "0.5"; + uint32_t precise_mode = 0; bool float_is_big_endian = false; // Set in arith_util.h by float_is_big_endian() tosa_level_t tosa_level; |