Notebook

In [1]:

%cd /workspace/axolotl

/workspace/axolotl

In [2]:

!accelerate config --config_file configs/accelerate/default_config.yaml default

Setting ds_accelerator to cuda (auto detect)
accelerate configuration saved at /root/.cache/huggingface/accelerate/default_config.yaml

In [4]:

!cat examples/falcon/config-1b-lora.yml

base_model: tiiuae/falcon-rw-1b
base_model_config: tiiuae/falcon-rw-1b
trust_remote_code: true
model_type: AutoModelForCausalLM
tokenizer_type: AutoTokenizer
load_in_8bit: true
load_in_4bit: false
gptq: false
strict: false
push_dataset_to_hub:
datasets:
  - path: teknium/GPT4-LLM-Cleaned
    type: alpaca:chat
dataset_prepared_path: last_run_prepared
val_set_size: 0.01
adapter: lora
lora_model_dir:
sequence_len: 2048
max_packed_sequence_len:
lora_r: 16
lora_alpha: 32
lora_dropout: 0.0
lora_target_modules:
lora_target_linear: true
lora_fan_in_fan_out:
wandb_project: falcon-rw-1b
wandb_watch:
wandb_run_id:
wandb_log_model:
output_dir: ./falcon-rw-1b
batch_size: 2
micro_batch_size: 1
num_epochs: 4
optimizer: adamw_bnb_8bit
torchdistx_path:
lr_scheduler: cosine
learning_rate: 0.00003
train_on_inputs: false
group_by_length: false
bf16: true
fp16: false
tf32: true
gradient_checkpointing: true
early_stopping_patience:
resume_from_checkpoint:
local_rank:
logging_steps: 1
xformers_attention: true
flash_attention:
gptq_groupsize:
gptq_model_v1:
warmup_steps: 40
eval_steps: 5
save_steps: 43
debug:
deepspeed:
weight_decay: 0.0
fsdp:
fsdp_config:
special_tokens:
  pad_token: "<|endoftext|>"
  bos_token: ">>ABSTRACT<<"
  eos_token: "<|endoftext|>"

In [6]:

!accelerate launch scripts/finetune.py examples/falcon/config-1b-lora.yml

Setting ds_accelerator to cuda (auto detect)

===================================BUG REPORT===================================
Welcome to bitsandbytes. For bug reports, please run

python -m bitsandbytes

 and submit this information together with your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
================================================================================
bin /root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/libbitsandbytes_cuda118.so
False
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/nvidia/lib'), PosixPath('/usr/local/nvidia/lib64')}
  warn(msg)
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: /usr/local/nvidia/lib:/usr/local/nvidia/lib64 did not contain ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] as expected! Searching further paths...
  warn(msg)
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('module'), PosixPath('//matplotlib_inline.backend_inline')}
  warn(msg)
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIK1tFOFrWbmoa2ckCJYhzgBHKTSMeR/AeuScCCzugqlI utensilcandel@gmail.com')}
  warn(msg)
CUDA_SETUP: WARNING! libcudart.so not found in any environmental path. Searching in backup paths...
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: Found duplicate ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] files: {PosixPath('/usr/local/cuda/lib64/libcudart.so'), PosixPath('/usr/local/cuda/lib64/libcudart.so.11.0')}.. We'll flip a coin and try one of these, in order to fail forward.
Either way, this might cause trouble in the future:
If you get `CUDA error: invalid device function` errors, the above might be the cause and the solution is to make sure only one ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] in the paths that we search based on your env.
  warn(msg)
CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so
CUDA SETUP: Highest compute capability among GPUs detected: 8.6
CUDA SETUP: Detected CUDA version 118
CUDA SETUP: Required library version not found: libbitsandbytes_cuda118.so. Maybe you need to compile it from source?
CUDA SETUP: Defaulting to libbitsandbytes_cpu.so...

================================================ERROR=====================================
CUDA SETUP: CUDA detection failed! Possible reasons:
1. CUDA driver not installed
2. CUDA not installed
3. You have multiple conflicting CUDA libraries
4. Required library not pre-compiled for this bitsandbytes release!
CUDA SETUP: If you compiled from source, try again with `make CUDA_VERSION=DETECTED_CUDA_VERSION` for example, `make CUDA_VERSION=113`.
CUDA SETUP: The CUDA version for the compile might depend on your conda install. Inspect CUDA version via `conda list | grep cuda`.
================================================================================

CUDA SETUP: Something unexpected happened. Please compile from source:
git clone git@github.com:TimDettmers/bitsandbytes.git
cd bitsandbytes
CUDA_VERSION=118 make cuda11x
python setup.py install
CUDA SETUP: Setup Failed!
Traceback (most recent call last):
  File "/workspace/axolotl/scripts/finetune.py", line 24, in <module>
    from axolotl.utils.models import load_model, load_tokenizer
  File "/workspace/axolotl/src/axolotl/utils/models.py", line 7, in <module>
    import bitsandbytes as bnb
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/__init__.py", line 6, in <module>
    from . import cuda_setup, utils, research
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/research/__init__.py", line 1, in <module>
    from . import nn
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/research/nn/__init__.py", line 1, in <module>
    from .modules import LinearFP8Mixed, LinearFP8Global
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/research/nn/modules.py", line 8, in <module>
    from bitsandbytes.optim import GlobalOptimManager
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/optim/__init__.py", line 6, in <module>
    from bitsandbytes.cextension import COMPILED_WITH_CUDA
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cextension.py", line 20, in <module>
    raise RuntimeError('''
RuntimeError: 
        CUDA Setup failed despite GPU being available. Please run the following command to get more information:

        python -m bitsandbytes

        Inspect the output of the command and see if you can locate CUDA libraries. You might need to add them
        to your LD_LIBRARY_PATH. If you suspect a bug, please take the information from python -m bitsandbytes
        and open an issue at: https://github.com/TimDettmers/bitsandbytes/issues
Traceback (most recent call last):
  File "/root/miniconda3/envs/py3.9/bin/accelerate", line 8, in <module>
    sys.exit(main())
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/accelerate/commands/accelerate_cli.py", line 45, in main
    args.func(args)
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/accelerate/commands/launch.py", line 928, in launch_command
    simple_launcher(args)
  File "/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/accelerate/commands/launch.py", line 588, in simple_launcher
    raise subprocess.CalledProcessError(returncode=process.returncode, cmd=cmd)
subprocess.CalledProcessError: Command '['/root/miniconda3/envs/py3.9/bin/python3', 'scripts/finetune.py', 'examples/falcon/config-1b-lora.yml']' returned non-zero exit status 1.

In [7]:

%cd /workspace

/workspace

In [8]:

!git clone https://github.com/TimDettmers/bitsandbytes

Cloning into 'bitsandbytes'...
remote: Enumerating objects: 5251, done.
remote: Counting objects: 100% (5250/5250), done.
remote: Compressing objects: 100% (1517/1517), done.
remote: Total 5251 (delta 3717), reused 5169 (delta 3684), pack-reused 1
Receiving objects: 100% (5251/5251), 1.48 MiB | 7.54 MiB/s, done.
Resolving deltas: 100% (3717/3717), done.

In [9]:

%cd bitsandbytes

/workspace/bitsandbytes

In [10]:

!CUDA_VERSION=118 make cuda11x
!python setup.py install
!python -m bitsandbytes

mkdir -p build
mkdir -p dependencies
ENVIRONMENT
============================
CUDA_VERSION: 118
============================
NVCC path: /usr/local/cuda/bin/nvcc
GPP path: /usr/bin/g++ VERSION: g++ (Ubuntu 11.3.0-1ubuntu1~22.04) 11.3.0
CUDA_HOME: /usr/local/cuda
CONDA_PREFIX: 
PATH: /root/miniconda3/envs/py3.9/bin:/root/miniconda3/bin:/usr/local/nvidia/bin:/usr/local/cuda/bin:/usr/local/sbin:/usr/local/bin:/usr/sbin:/usr/bin:/sbin:/bin
LD_LIBRARY_PATH: /usr/local/nvidia/lib:/usr/local/nvidia/lib64
============================
/usr/local/cuda/bin/nvcc -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -Xcompiler '-fPIC' --use_fast_math -Xptxas=-v -dc /workspace/bitsandbytes/csrc/ops.cu /workspace/bitsandbytes/csrc/kernels.cu -I /usr/local/cuda/include -I /workspace/bitsandbytes/csrc -I /include -I /workspace/bitsandbytes/include -L /usr/local/cuda/lib64 -lcudart -lcublas -lcublasLt -lcusparse -L /lib --output-directory /workspace/bitsandbytes/build
ptxas info    : 15 bytes gmem
ptxas info    : Compiling entry function '_ZN3cub11EmptyKernelIvEEvv' for 'sm_75'
ptxas info    : Function properties for _ZN3cub11EmptyKernelIvEEvv
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 4 registers, 352 bytes cmem[0]
ptxas info    : 15 bytes gmem
ptxas info    : Compiling entry function '_ZN3cub11EmptyKernelIvEEvv' for 'sm_80'
ptxas info    : Function properties for _ZN3cub11EmptyKernelIvEEvv
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 4 registers, 352 bytes cmem[0]
ptxas info    : 15 bytes gmem
ptxas info    : Compiling entry function '_ZN3cub11EmptyKernelIvEEvv' for 'sm_86'
ptxas info    : Function properties for _ZN3cub11EmptyKernelIvEEvv
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 4 registers, 352 bytes cmem[0]
ptxas warning : Value of threads per SM for entry _Z9kQuantizePfS_Phi is out of range. .minnctapersm will be ignored
ptxas info    : 31 bytes gmem
ptxas info    : Compiling entry function '_ZN3cub11EmptyKernelIvEEvv' for 'sm_75'
ptxas info    : Function properties for _ZN3cub11EmptyKernelIvEEvv
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 4 registers, 352 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi4ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi4ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi4ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi4ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI13__nv_bfloat16Li5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI13__nv_bfloat16Li5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi5ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi5ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi2ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi2ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi2ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi2ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi1ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi1ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi1ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi1ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 74 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseI13__nv_bfloat16Li0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseI13__nv_bfloat16Li0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseI6__halfLi0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseI6__halfLi0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseIfLi0ELi2048ELi8EEvPT_S1_PhS2_fffifPfS3_S3_S3_ffbi' for 'sm_75'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseIfLi0ELi2048ELi8EEvPT_S1_PhS2_fffifPfS3_S3_S3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi2EEvPfPhS0_PT_ii' for 'sm_75'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi2EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi2EEvPfPhS1_PT_ii' for 'sm_75'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi2EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi0EEvPfPhS0_PT_ii' for 'sm_75'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi0EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi0EEvPfPhS1_PT_ii' for 'sm_75'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi0EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 59 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi1EEvPfPhS0_PT_ii' for 'sm_75'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi1EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi1EEvPfPhS1_PT_ii' for 'sm_75'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi1EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 29 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 29 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 38 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 45 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 45 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 45 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 51 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 51 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi1ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi1ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 57 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 51 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 44 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 44 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 44 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 50 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 50 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi1ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi1ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 59 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_75'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 50 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z19kPercentileClippingI6__halfLi2048ELi4EEvPT_Pfii' for 'sm_75'
ptxas info    : Function properties for _Z19kPercentileClippingI6__halfLi2048ELi4EEvPT_Pfii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 35 registers, 376 bytes cmem[0]
ptxas info    : Compiling entry function '_Z19kPercentileClippingIfLi2048ELi4EEvPT_Pfii' for 'sm_75'
ptxas info    : Function properties for _Z19kPercentileClippingIfLi2048ELi4EEvPT_Pfii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 35 registers, 376 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PKffffffifPfS5_S5_S5_S5_S5_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PKffffffifPfS5_S5_S5_S5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 484 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PKffffffifPfS6_S6_S6_S6_S6_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PKffffffifPfS6_S6_S6_S6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 484 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PffffiS3_S3_S3_S3_S3_S3_fi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PffffiS3_S3_S3_S3_S3_S3_fi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 115 registers, 464 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PffffiS4_S4_S4_S4_S4_S4_fi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PffffiS4_S4_S4_S4_S4_S4_fi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 115 registers, 464 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_75'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_75'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateI13__nv_bfloat16Li0EEvPT_S2_PfS3_S3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateI13__nv_bfloat16Li0EEvPT_S2_PfS3_S3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateI6__halfLi0EEvPT_S2_PfS3_S3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateI6__halfLi0EEvPT_S2_PfS3_S3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateIfLi0EEvPT_S1_PfS2_S2_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateIfLi0EEvPT_S1_PfS2_S2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateI13__nv_bfloat16Li0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateI13__nv_bfloat16Li0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 57 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateI6__halfLi0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateI6__halfLi0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 57 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateIfLi0ELi4096ELi8EEvPT_S1_PfS2_S2_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateIfLi0ELi4096ELi8EEvPT_S1_PfS2_S2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 55 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi4EEvPT_S1_PfS2_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi4EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi4EEvPT_S2_PfS3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi4EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI13__nv_bfloat16Li5EEvPT_S2_PfS3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI13__nv_bfloat16Li5EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 49 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi5EEvPT_S1_PfS2_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi5EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 49 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi5EEvPT_S2_PfS3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi5EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 50 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi2EEvPT_S1_PfS2_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi2EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 49 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi2EEvPT_S2_PfS3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi2EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 50 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi1EEvPT_S1_PfS2_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi1EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi1EEvPT_S2_PfS3_ffffffiffbi' for 'sm_75'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi1EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 50 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi4ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi4ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi4ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi4ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 53 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI13__nv_bfloat16Li5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI13__nv_bfloat16Li5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi5ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi5ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 53 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi2ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi2ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi2ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi2ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 53 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi1ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi1ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi1ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_75'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi1ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 43 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kEstimateQuantilesI6__halfEvPT_PffS1_i' for 'sm_75'
ptxas info    : Function properties for _Z18kEstimateQuantilesI6__halfEvPT_PffS1_i
    16 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 82 registers, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kEstimateQuantilesIfEvPT_PffS0_i' for 'sm_75'
ptxas info    : Function properties for _Z18kEstimateQuantilesIfEvPT_PffS0_i
    32 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 84 registers, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi1EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii' for 'sm_75'
ptxas info    : Function properties for _Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi1EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 38 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi0EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii' for 'sm_75'
ptxas info    : Function properties for _Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi0EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z22kdequant_mm_int32_fp16ILi4ELi128ELi512EEvPiPfS1_P6__halfS1_S1_S3_iiii' for 'sm_75'
ptxas info    : Function properties for _Z22kdequant_mm_int32_fp16ILi4ELi128ELi512EEvPiPfS1_P6__halfS1_S1_S3_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 42 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi4EEvPcS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi4EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi4EEvPcS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi4EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 42 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi3EEvPcS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi3EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi3EEvPcS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi3EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 43 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi2EEvPcS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi2EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 35 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi2EEvPcS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi2EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi32ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_75'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi32ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi16ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_75'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi16ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi8ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_75'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi8ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi32ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_75'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi32ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi16ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_75'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi16ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi8ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_75'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi8ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z16kExtractOutliersILi4EEvPcPiS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z16kExtractOutliersILi4EEvPcPiS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 396 bytes cmem[0]
ptxas info    : Compiling entry function '_Z16kExtractOutliersILi3EEvPcPiS0_iiiii' for 'sm_75'
ptxas info    : Function properties for _Z16kExtractOutliersILi3EEvPcPiS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 396 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kgemm_4bit_inferenceI6__halfLi160EEviiiPT_PhPfS2_iiii' for 'sm_75'
ptxas info    : Function properties for _Z20kgemm_4bit_inferenceI6__halfLi160EEviiiPT_PhPfS2_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 416 bytes cmem[0], 48 bytes cmem[2]
ptxas info    : Compiling entry function '_Z20kgemm_4bit_inferenceI6__halfLi128EEviiiPT_PhPfS2_iiii' for 'sm_75'
ptxas info    : Function properties for _Z20kgemm_4bit_inferenceI6__halfLi128EEviiiPT_PhPfS2_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 416 bytes cmem[0], 48 bytes cmem[2]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi96EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi96EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi64EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi64EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi32EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi32EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi128EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi128EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi160EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi160EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi192EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi192EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi256EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi256EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi96EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi96EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi64EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi64EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi32EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi32EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi128EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi128EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi160EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi160EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi192EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi192EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi256EEviiiPT_S2_S2_iii' for 'sm_75'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi256EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi2EEvPT_S1_S0_l' for 'sm_75'
ptxas info    : Function properties for _Z5kfuncIfLi2EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi1EEvPT_S1_S0_l' for 'sm_75'
ptxas info    : Function properties for _Z5kfuncIfLi1EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIhLi0EEvPT_S1_S0_l' for 'sm_75'
ptxas info    : Function properties for _Z5kfuncIhLi0EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi0EEvPT_S1_S0_l' for 'sm_75'
ptxas info    : Function properties for _Z5kfuncIfLi0EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi1EEvPT_PfS3_Pifiiii' for 'sm_75'
ptxas info    : Function properties for _Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi1EEvPT_PfS3_Pifiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi0EEvPT_PfS3_Pifiiii' for 'sm_75'
ptxas info    : Function properties for _Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi0EEvPT_PfS3_Pifiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11kDequantizePfPhS_i' for 'sm_75'
ptxas info    : Function properties for _Z11kDequantizePfPhS_i
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 12 registers, 1024 bytes smem, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z9kQuantizePfS_Phi' for 'sm_75'
ptxas info    : Function properties for _Z9kQuantizePfS_Phi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 52 registers, 21520 bytes smem, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z22kHistogramScatterAdd2DPfPiS0_S_ii' for 'sm_75'
ptxas info    : Function properties for _Z22kHistogramScatterAdd2DPfPiS0_S_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 392 bytes cmem[0]
ptxas info    : Function properties for _Z9dQuantizeILi1EEhPfff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9dQuantizeILi0EEhPfff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z12dQuantizeNF4f
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z14dDequantizeNF4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z15dhDequantizeNF4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z12dQuantizeFP4f
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z18dDequantizeFP4Treehf
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z15d2DequantizeFP4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z14dDequantizeFP4hf
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9atomicMinPff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9atomicMaxPff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas warning : Value of threads per SM for entry _Z9kQuantizePfS_Phi is out of range. .minnctapersm will be ignored
ptxas info    : 31 bytes gmem
ptxas info    : Compiling entry function '_ZN3cub11EmptyKernelIvEEvv' for 'sm_80'
ptxas info    : Function properties for _ZN3cub11EmptyKernelIvEEvv
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 4 registers, 352 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi4ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi4ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi4ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi4ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI13__nv_bfloat16Li5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI13__nv_bfloat16Li5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi5ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi5ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi2ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi2ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi2ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi2ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi1ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi1ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi1ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi1ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 74 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseI13__nv_bfloat16Li0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseI13__nv_bfloat16Li0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseI6__halfLi0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseI6__halfLi0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseIfLi0ELi2048ELi8EEvPT_S1_PhS2_fffifPfS3_S3_S3_ffbi' for 'sm_80'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseIfLi0ELi2048ELi8EEvPT_S1_PhS2_fffifPfS3_S3_S3_ffbi
    8 bytes stack frame, 4 bytes spill stores, 4 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi2EEvPfPhS0_PT_ii' for 'sm_80'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi2EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 56 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi2EEvPfPhS1_PT_ii' for 'sm_80'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi2EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 56 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi0EEvPfPhS0_PT_ii' for 'sm_80'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi0EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi0EEvPfPhS1_PT_ii' for 'sm_80'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi0EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi1EEvPfPhS0_PT_ii' for 'sm_80'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi1EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi1EEvPfPhS1_PT_ii' for 'sm_80'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi1EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi1ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi1ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi1ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi1ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_80'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z19kPercentileClippingI6__halfLi2048ELi4EEvPT_Pfii' for 'sm_80'
ptxas info    : Function properties for _Z19kPercentileClippingI6__halfLi2048ELi4EEvPT_Pfii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 376 bytes cmem[0]
ptxas info    : Compiling entry function '_Z19kPercentileClippingIfLi2048ELi4EEvPT_Pfii' for 'sm_80'
ptxas info    : Function properties for _Z19kPercentileClippingIfLi2048ELi4EEvPT_Pfii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 376 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PKffffffifPfS5_S5_S5_S5_S5_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PKffffffifPfS5_S5_S5_S5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 484 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PKffffffifPfS6_S6_S6_S6_S6_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PKffffffifPfS6_S6_S6_S6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 484 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PffffiS3_S3_S3_S3_S3_S3_fi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PffffiS3_S3_S3_S3_S3_S3_fi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 115 registers, 464 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PffffiS4_S4_S4_S4_S4_S4_fi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PffffiS4_S4_S4_S4_S4_S4_fi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 120 registers, 464 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_80'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_80'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateI13__nv_bfloat16Li0EEvPT_S2_PfS3_S3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateI13__nv_bfloat16Li0EEvPT_S2_PfS3_S3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateI6__halfLi0EEvPT_S2_PfS3_S3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateI6__halfLi0EEvPT_S2_PfS3_S3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateIfLi0EEvPT_S1_PfS2_S2_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateIfLi0EEvPT_S1_PfS2_S2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateI13__nv_bfloat16Li0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateI13__nv_bfloat16Li0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 55 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateI6__halfLi0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateI6__halfLi0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 55 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateIfLi0ELi4096ELi8EEvPT_S1_PfS2_S2_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateIfLi0ELi4096ELi8EEvPT_S1_PfS2_S2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 56 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi4EEvPT_S1_PfS2_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi4EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi4EEvPT_S2_PfS3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi4EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI13__nv_bfloat16Li5EEvPT_S2_PfS3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI13__nv_bfloat16Li5EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi5EEvPT_S1_PfS2_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi5EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi5EEvPT_S2_PfS3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi5EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi2EEvPT_S1_PfS2_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi2EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi2EEvPT_S2_PfS3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi2EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi1EEvPT_S1_PfS2_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi1EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi1EEvPT_S2_PfS3_ffffffiffbi' for 'sm_80'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi1EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi4ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi4ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi4ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi4ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI13__nv_bfloat16Li5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI13__nv_bfloat16Li5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi5ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi5ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi2ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi2ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi2ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi2ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi1ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi1ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi1ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_80'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi1ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kEstimateQuantilesI6__halfEvPT_PffS1_i' for 'sm_80'
ptxas info    : Function properties for _Z18kEstimateQuantilesI6__halfEvPT_PffS1_i
    16 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 82 registers, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kEstimateQuantilesIfEvPT_PffS0_i' for 'sm_80'
ptxas info    : Function properties for _Z18kEstimateQuantilesIfEvPT_PffS0_i
    32 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 83 registers, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi1EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii' for 'sm_80'
ptxas info    : Function properties for _Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi1EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi0EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii' for 'sm_80'
ptxas info    : Function properties for _Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi0EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z22kdequant_mm_int32_fp16ILi4ELi128ELi512EEvPiPfS1_P6__halfS1_S1_S3_iiii' for 'sm_80'
ptxas info    : Function properties for _Z22kdequant_mm_int32_fp16ILi4ELi128ELi512EEvPiPfS1_P6__halfS1_S1_S3_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi4EEvPcS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi4EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi4EEvPcS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi4EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi3EEvPcS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi3EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi3EEvPcS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi3EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi2EEvPcS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi2EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi2EEvPcS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi2EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi32ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_80'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi32ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi16ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_80'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi16ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi8ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_80'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi8ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi32ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_80'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi32ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi16ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_80'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi16ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi8ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_80'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi8ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 32 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z16kExtractOutliersILi4EEvPcPiS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z16kExtractOutliersILi4EEvPcPiS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 396 bytes cmem[0]
ptxas info    : Compiling entry function '_Z16kExtractOutliersILi3EEvPcPiS0_iiiii' for 'sm_80'
ptxas info    : Function properties for _Z16kExtractOutliersILi3EEvPcPiS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 13 registers, 396 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kgemm_4bit_inferenceI6__halfLi160EEviiiPT_PhPfS2_iiii' for 'sm_80'
ptxas info    : Function properties for _Z20kgemm_4bit_inferenceI6__halfLi160EEviiiPT_PhPfS2_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kgemm_4bit_inferenceI6__halfLi128EEviiiPT_PhPfS2_iiii' for 'sm_80'
ptxas info    : Function properties for _Z20kgemm_4bit_inferenceI6__halfLi128EEviiiPT_PhPfS2_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi96EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi96EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi64EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi64EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi32EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi32EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi128EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi128EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi160EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi160EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi192EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi192EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi256EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi256EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi96EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi96EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi64EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi64EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi32EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi32EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi128EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi128EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi160EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi160EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi192EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi192EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi256EEviiiPT_S2_S2_iii' for 'sm_80'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi256EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 167 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi2EEvPT_S1_S0_l' for 'sm_80'
ptxas info    : Function properties for _Z5kfuncIfLi2EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 26 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi1EEvPT_S1_S0_l' for 'sm_80'
ptxas info    : Function properties for _Z5kfuncIfLi1EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIhLi0EEvPT_S1_S0_l' for 'sm_80'
ptxas info    : Function properties for _Z5kfuncIhLi0EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi0EEvPT_S1_S0_l' for 'sm_80'
ptxas info    : Function properties for _Z5kfuncIfLi0EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi1EEvPT_PfS3_Pifiiii' for 'sm_80'
ptxas info    : Function properties for _Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi1EEvPT_PfS3_Pifiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 29 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi0EEvPT_PfS3_Pifiiii' for 'sm_80'
ptxas info    : Function properties for _Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi0EEvPT_PfS3_Pifiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11kDequantizePfPhS_i' for 'sm_80'
ptxas info    : Function properties for _Z11kDequantizePfPhS_i
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 12 registers, 1024 bytes smem, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z9kQuantizePfS_Phi' for 'sm_80'
ptxas info    : Function properties for _Z9kQuantizePfS_Phi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 51 registers, 21520 bytes smem, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z22kHistogramScatterAdd2DPfPiS0_S_ii' for 'sm_80'
ptxas info    : Function properties for _Z22kHistogramScatterAdd2DPfPiS0_S_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 392 bytes cmem[0]
ptxas info    : Function properties for _Z9dQuantizeILi1EEhPfff
    8 bytes stack frame, 4 bytes spill stores, 4 bytes spill loads
ptxas info    : Function properties for _Z9dQuantizeILi0EEhPfff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z12dQuantizeNF4f
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z14dDequantizeNF4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z15dhDequantizeNF4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z12dQuantizeFP4f
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z18dDequantizeFP4Treehf
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z15d2DequantizeFP4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z14dDequantizeFP4hf
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9atomicMinPff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9atomicMaxPff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas warning : Value of threads per SM for entry _Z9kQuantizePfS_Phi is out of range. .minnctapersm will be ignored
ptxas info    : 31 bytes gmem
ptxas info    : Compiling entry function '_ZN3cub11EmptyKernelIvEEvv' for 'sm_86'
ptxas info    : Function properties for _ZN3cub11EmptyKernelIvEEvv
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 4 registers, 352 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi4ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi4ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi4ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi4ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI13__nv_bfloat16Li5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI13__nv_bfloat16Li5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi5ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi5ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi5ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi2ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi2ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi2ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi2ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi1ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseI6__halfLi1ELi2048ELi8EEvPT_S2_PhfffifPfS4_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit1StateBlockwiseIfLi1ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit1StateBlockwiseIfLi1ELi2048ELi8EEvPT_S1_PhfffifPfS3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 74 registers, 432 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseI13__nv_bfloat16Li0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseI13__nv_bfloat16Li0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi
    16 bytes stack frame, 16 bytes spill stores, 16 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseI6__halfLi0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseI6__halfLi0ELi2048ELi8EEvPT_S2_PhS3_fffifPfS4_S4_S4_ffbi
    16 bytes stack frame, 16 bytes spill stores, 16 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z35kOptimizerStatic8bit2StateBlockwiseIfLi0ELi2048ELi8EEvPT_S1_PhS2_fffifPfS3_S3_S3_ffbi' for 'sm_86'
ptxas info    : Function properties for _Z35kOptimizerStatic8bit2StateBlockwiseIfLi0ELi2048ELi8EEvPT_S1_PhS2_fffifPfS3_S3_S3_ffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 80 registers, 456 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi2EEvPfPhS0_PT_ii' for 'sm_86'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi2EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi2EEvPfPhS1_PT_ii' for 'sm_86'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi2EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi0EEvPfPhS0_PT_ii' for 'sm_86'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi0EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi0EEvPfPhS1_PT_ii' for 'sm_86'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi0EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi1EEvPfPhS0_PT_ii' for 'sm_86'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseIfLi512ELi64ELi8ELi1EEvPfPhS0_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi1EEvPfPhS1_PT_ii' for 'sm_86'
ptxas info    : Function properties for _Z20kDequantizeBlockwiseI6__halfLi512ELi64ELi8ELi1EEvPfPhS1_PT_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 392 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi2EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi2EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi1EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 27 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi1EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 34 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi64ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi128ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi256ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi512ELi2ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi1024ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi2048ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi1ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi1ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseIfLi4096ELi4ELi0ELi0EEvPfPT_S0_PhS0_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi64ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 38 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi128ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi256ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi512ELi2ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi1024ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi2048ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi1ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi1ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii' for 'sm_86'
ptxas info    : Function properties for _Z18kQuantizeBlockwiseI6__halfLi4096ELi4ELi0ELi0EEvPfPT_S1_PhS1_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 400 bytes cmem[0]
ptxas info    : Compiling entry function '_Z19kPercentileClippingI6__halfLi2048ELi4EEvPT_Pfii' for 'sm_86'
ptxas info    : Function properties for _Z19kPercentileClippingI6__halfLi2048ELi4EEvPT_Pfii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 37 registers, 376 bytes cmem[0]
ptxas info    : Compiling entry function '_Z19kPercentileClippingIfLi2048ELi4EEvPT_Pfii' for 'sm_86'
ptxas info    : Function properties for _Z19kPercentileClippingIfLi2048ELi4EEvPT_Pfii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 39 registers, 376 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PKffffffifPfS5_S5_S5_S5_S5_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PKffffffifPfS5_S5_S5_S5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 484 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PKffffffifPfS6_S6_S6_S6_S6_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PKffffffifPfS6_S6_S6_S6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 484 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PffffiS3_S3_S3_S3_S3_S3_fi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit2StateIfLi0EEvPT_S1_PhS2_PffffiS3_S3_S3_S3_S3_S3_fi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 115 registers, 464 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PffffiS4_S4_S4_S4_S4_S4_fi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit2StateI6__halfLi0EEvPT_S2_PhS3_PffffiS4_S4_S4_S4_S4_S4_fi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 120 registers, 464 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 62 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPKffffffifPfS5_S5_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPKffffffifPfS5_S5_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z26kOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPKffffffifPfS6_S6_ffi' for 'sm_86'
ptxas info    : Function properties for _Z26kOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPKffffffifPfS6_S6_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 62 registers, 452 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi5EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi5EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi2EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi2EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPffffiS3_S3_S3_ffi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateIfLi1EEvPT_S1_PhPffffiS3_S3_S3_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPffffiS4_S4_S4_ffi' for 'sm_86'
ptxas info    : Function properties for _Z38kPreconditionOptimizerStatic8bit1StateI6__halfLi1EEvPT_S2_PhPffffiS4_S4_S4_ffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateI13__nv_bfloat16Li0EEvPT_S2_PfS3_S3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateI13__nv_bfloat16Li0EEvPT_S2_PfS3_S3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateI6__halfLi0EEvPT_S2_PfS3_S3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateI6__halfLi0EEvPT_S2_PfS3_S3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 64 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit2StateIfLi0EEvPT_S1_PfS2_S2_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit2StateIfLi0EEvPT_S1_PfS2_S2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 63 registers, 436 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateI13__nv_bfloat16Li0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateI13__nv_bfloat16Li0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 55 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateI6__halfLi0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateI6__halfLi0ELi4096ELi8EEvPT_S2_PfS3_S3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 55 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit2StateIfLi0ELi4096ELi8EEvPT_S1_PfS2_S2_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit2StateIfLi0ELi4096ELi8EEvPT_S1_PfS2_S2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 56 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi4EEvPT_S1_PfS2_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi4EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi4EEvPT_S2_PfS3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi4EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI13__nv_bfloat16Li5EEvPT_S2_PfS3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI13__nv_bfloat16Li5EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi5EEvPT_S1_PfS2_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi5EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi5EEvPT_S2_PfS3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi5EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi2EEvPT_S1_PfS2_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi2EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi2EEvPT_S2_PfS3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi2EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateIfLi1EEvPT_S1_PfS2_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateIfLi1EEvPT_S1_PfS2_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kOptimizer32bit1StateI6__halfLi1EEvPT_S2_PfS3_ffffffiffbi' for 'sm_86'
ptxas info    : Function properties for _Z21kOptimizer32bit1StateI6__halfLi1EEvPT_S2_PfS3_ffffffiffbi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 428 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi4ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi4ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi4ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi4ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI13__nv_bfloat16Li5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI13__nv_bfloat16Li5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi5ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi5ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 48 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi5ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi2ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi2ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi2ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi2ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateIfLi1ELi4096ELi8EEvPT_S1_PfS2_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateIfLi1ELi4096ELi8EEvPT_S1_PfS2_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 47 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z33kPreconditionOptimizer32bit1StateI6__halfLi1ELi4096ELi8EEvPT_S2_PfS3_ffffiffi' for 'sm_86'
ptxas info    : Function properties for _Z33kPreconditionOptimizer32bit1StateI6__halfLi1ELi4096ELi8EEvPT_S2_PfS3_ffffiffi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 46 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kEstimateQuantilesI6__halfEvPT_PffS1_i' for 'sm_86'
ptxas info    : Function properties for _Z18kEstimateQuantilesI6__halfEvPT_PffS1_i
    16 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 82 registers, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kEstimateQuantilesIfEvPT_PffS0_i' for 'sm_86'
ptxas info    : Function properties for _Z18kEstimateQuantilesIfEvPT_PffS0_i
    32 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 83 registers, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi1EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii' for 'sm_86'
ptxas info    : Function properties for _Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi1EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 38 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi0EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii' for 'sm_86'
ptxas info    : Function properties for _Z18kDoubleRowColQuantILi64ELi4ELi16ELi256ELi0EEvP6__halfPfS2_PcS3_PiS4_S1_S4_fiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 36 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z22kdequant_mm_int32_fp16ILi4ELi128ELi512EEvPiPfS1_P6__halfS1_S1_S3_iiii' for 'sm_86'
ptxas info    : Function properties for _Z22kdequant_mm_int32_fp16ILi4ELi128ELi512EEvPiPfS1_P6__halfS1_S1_S3_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 37 registers, 424 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi4EEvPcS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi4EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 31 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi4EEvPcS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi4EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi3EEvPcS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi3EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi3EEvPcS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi3EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi2EEvPcS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi1ELi2EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi2EEvPcS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z21kTransformRowToFormatILi256ELi8ELi32ELi256ELi0ELi2EEvPcS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 388 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi32ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_86'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi32ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi16ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_86'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi16ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveIaLi8ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii' for 'sm_86'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveIaLi8ELi8EEvPiS0_S0_S0_S0_P6__halfPT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi32ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_86'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi32ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi16ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_86'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi16ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z27kspmm_coo_very_sparse_naiveI6__halfLi8ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii' for 'sm_86'
ptxas info    : Function properties for _Z27kspmm_coo_very_sparse_naiveI6__halfLi8ELi16EEvPiS1_S1_S1_S1_PS0_PT_S2_Pfiiii
    192 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 40 registers, 440 bytes cmem[0]
ptxas info    : Compiling entry function '_Z16kExtractOutliersILi4EEvPcPiS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z16kExtractOutliersILi4EEvPcPiS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 396 bytes cmem[0]
ptxas info    : Compiling entry function '_Z16kExtractOutliersILi3EEvPcPiS0_iiiii' for 'sm_86'
ptxas info    : Function properties for _Z16kExtractOutliersILi3EEvPcPiS0_iiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 13 registers, 396 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kgemm_4bit_inferenceI6__halfLi160EEviiiPT_PhPfS2_iiii' for 'sm_86'
ptxas info    : Function properties for _Z20kgemm_4bit_inferenceI6__halfLi160EEviiiPT_PhPfS2_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 72 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z20kgemm_4bit_inferenceI6__halfLi128EEviiiPT_PhPfS2_iiii' for 'sm_86'
ptxas info    : Function properties for _Z20kgemm_4bit_inferenceI6__halfLi128EEviiiPT_PhPfS2_iiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 72 registers, 416 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi96EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi96EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi64EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi64EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi32EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi32EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi128EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi128EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi160EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi160EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi192EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi192EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi16ELi256EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi16ELi256EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi96EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi96EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi64EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi64EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi32EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi32EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi128EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi128EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi160EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi160EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi192EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi192EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11gemm_deviceI6__halfLi32ELi256EEviiiPT_S2_S2_iii' for 'sm_86'
ptxas info    : Function properties for _Z11gemm_deviceI6__halfLi32ELi256EEviiiPT_S2_S2_iii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 168 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi2EEvPT_S1_S0_l' for 'sm_86'
ptxas info    : Function properties for _Z5kfuncIfLi2EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi1EEvPT_S1_S0_l' for 'sm_86'
ptxas info    : Function properties for _Z5kfuncIfLi1EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 30 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIhLi0EEvPT_S1_S0_l' for 'sm_86'
ptxas info    : Function properties for _Z5kfuncIhLi0EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z5kfuncIfLi0EEvPT_S1_S0_l' for 'sm_86'
ptxas info    : Function properties for _Z5kfuncIfLi0EEvPT_S1_S0_l
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 24 registers, 384 bytes cmem[0]
ptxas info    : Compiling entry function '_Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi1EEvPT_PfS3_Pifiiii' for 'sm_86'
ptxas info    : Function properties for _Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi1EEvPT_PfS3_Pifiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi0EEvPT_PfS3_Pifiiii' for 'sm_86'
ptxas info    : Function properties for _Z15kgetColRowStatsI6__halfLi64ELi4ELi16ELi256ELi0EEvPT_PfS3_Pifiiii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 28 registers, 404 bytes cmem[0]
ptxas info    : Compiling entry function '_Z11kDequantizePfPhS_i' for 'sm_86'
ptxas info    : Function properties for _Z11kDequantizePfPhS_i
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 12 registers, 1024 bytes smem, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z9kQuantizePfS_Phi' for 'sm_86'
ptxas info    : Function properties for _Z9kQuantizePfS_Phi
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 51 registers, 21520 bytes smem, 380 bytes cmem[0]
ptxas info    : Compiling entry function '_Z22kHistogramScatterAdd2DPfPiS0_S_ii' for 'sm_86'
ptxas info    : Function properties for _Z22kHistogramScatterAdd2DPfPiS0_S_ii
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Used 14 registers, 392 bytes cmem[0]
ptxas info    : Function properties for _Z9dQuantizeILi1EEhPfff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9dQuantizeILi0EEhPfff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z12dQuantizeNF4f
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z14dDequantizeNF4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z15dhDequantizeNF4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z12dQuantizeFP4f
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z18dDequantizeFP4Treehf
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z15d2DequantizeFP4h
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z14dDequantizeFP4hf
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9atomicMinPff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
ptxas info    : Function properties for _Z9atomicMaxPff
    0 bytes stack frame, 0 bytes spill stores, 0 bytes spill loads
/usr/local/cuda/bin/nvcc -gencode arch=compute_75,code=sm_75 -gencode arch=compute_80,code=sm_80 -gencode arch=compute_86,code=sm_86 -Xcompiler '-fPIC' -dlink /workspace/bitsandbytes/build/ops.o /workspace/bitsandbytes/build/kernels.o -o /workspace/bitsandbytes/build/link.o
/usr/bin/g++ -std=c++14 -DBUILD_CUDA -shared -fPIC -I /usr/local/cuda/include -I /workspace/bitsandbytes/csrc -I /include -I /workspace/bitsandbytes/include /workspace/bitsandbytes/build/ops.o /workspace/bitsandbytes/build/kernels.o /workspace/bitsandbytes/build/link.o /workspace/bitsandbytes/csrc/common.cpp /workspace/bitsandbytes/csrc/cpu_ops.cpp /workspace/bitsandbytes/csrc/pythonInterface.c -o ./bitsandbytes/libbitsandbytes_cuda118.so -L /usr/local/cuda/lib64 -lcudart -lcublas -lcublasLt -lcusparse -L /lib
libs: ['libbitsandbytes_cuda118.so']
running install
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/setuptools/command/install.py:34: SetuptoolsDeprecationWarning: setup.py install is deprecated. Use build and pip and other standards-based tools.
  warnings.warn(
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/setuptools/command/easy_install.py:144: EasyInstallDeprecationWarning: easy_install command is deprecated. Use build and pip and other standards-based tools.
  warnings.warn(
running bdist_egg
running egg_info
creating bitsandbytes.egg-info
writing bitsandbytes.egg-info/PKG-INFO
writing dependency_links to bitsandbytes.egg-info/dependency_links.txt
writing top-level names to bitsandbytes.egg-info/top_level.txt
writing manifest file 'bitsandbytes.egg-info/SOURCES.txt'
reading manifest file 'bitsandbytes.egg-info/SOURCES.txt'
adding license file 'LICENSE'
adding license file 'NOTICE.md'
writing manifest file 'bitsandbytes.egg-info/SOURCES.txt'
installing library code to build/bdist.linux-x86_64/egg
running install_lib
running build_py
creating build/lib
creating build/lib/bitsandbytes
copying bitsandbytes/__init__.py -> build/lib/bitsandbytes
copying bitsandbytes/__main__.py -> build/lib/bitsandbytes
copying bitsandbytes/cextension.py -> build/lib/bitsandbytes
copying bitsandbytes/functional.py -> build/lib/bitsandbytes
copying bitsandbytes/utils.py -> build/lib/bitsandbytes
creating build/lib/bitsandbytes/autograd
copying bitsandbytes/autograd/__init__.py -> build/lib/bitsandbytes/autograd
copying bitsandbytes/autograd/_functions.py -> build/lib/bitsandbytes/autograd
creating build/lib/bitsandbytes/cuda_setup
copying bitsandbytes/cuda_setup/__init__.py -> build/lib/bitsandbytes/cuda_setup
copying bitsandbytes/cuda_setup/env_vars.py -> build/lib/bitsandbytes/cuda_setup
copying bitsandbytes/cuda_setup/main.py -> build/lib/bitsandbytes/cuda_setup
creating build/lib/bitsandbytes/nn
copying bitsandbytes/nn/__init__.py -> build/lib/bitsandbytes/nn
copying bitsandbytes/nn/modules.py -> build/lib/bitsandbytes/nn
copying bitsandbytes/nn/triton_based_modules.py -> build/lib/bitsandbytes/nn
creating build/lib/bitsandbytes/optim
copying bitsandbytes/optim/__init__.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/adagrad.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/adam.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/adamw.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/lamb.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/lars.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/lion.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/optimizer.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/rmsprop.py -> build/lib/bitsandbytes/optim
copying bitsandbytes/optim/sgd.py -> build/lib/bitsandbytes/optim
creating build/lib/bitsandbytes/research
copying bitsandbytes/research/__init__.py -> build/lib/bitsandbytes/research
creating build/lib/bitsandbytes/triton
copying bitsandbytes/triton/__init__.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/dequantize_rowwise.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/int8_matmul_mixed_dequanitze.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/int8_matmul_rowwise_dequantize.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/quantize_columnwise_and_transpose.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/quantize_global.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/quantize_rowwise.py -> build/lib/bitsandbytes/triton
copying bitsandbytes/triton/triton_utils.py -> build/lib/bitsandbytes/triton
creating build/lib/bitsandbytes/research/autograd
copying bitsandbytes/research/autograd/__init__.py -> build/lib/bitsandbytes/research/autograd
copying bitsandbytes/research/autograd/_functions.py -> build/lib/bitsandbytes/research/autograd
creating build/lib/bitsandbytes/research/nn
copying bitsandbytes/research/nn/__init__.py -> build/lib/bitsandbytes/research/nn
copying bitsandbytes/research/nn/modules.py -> build/lib/bitsandbytes/research/nn
copying bitsandbytes/libbitsandbytes_cuda118.so -> build/lib/bitsandbytes
creating build/bdist.linux-x86_64
creating build/bdist.linux-x86_64/egg
creating build/bdist.linux-x86_64/egg/bitsandbytes
copying build/lib/bitsandbytes/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes
copying build/lib/bitsandbytes/__main__.py -> build/bdist.linux-x86_64/egg/bitsandbytes
copying build/lib/bitsandbytes/cextension.py -> build/bdist.linux-x86_64/egg/bitsandbytes
copying build/lib/bitsandbytes/functional.py -> build/bdist.linux-x86_64/egg/bitsandbytes
copying build/lib/bitsandbytes/utils.py -> build/bdist.linux-x86_64/egg/bitsandbytes
creating build/bdist.linux-x86_64/egg/bitsandbytes/autograd
copying build/lib/bitsandbytes/autograd/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/autograd
copying build/lib/bitsandbytes/autograd/_functions.py -> build/bdist.linux-x86_64/egg/bitsandbytes/autograd
creating build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup
copying build/lib/bitsandbytes/cuda_setup/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup
copying build/lib/bitsandbytes/cuda_setup/env_vars.py -> build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup
copying build/lib/bitsandbytes/cuda_setup/main.py -> build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup
creating build/bdist.linux-x86_64/egg/bitsandbytes/nn
copying build/lib/bitsandbytes/nn/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/nn
copying build/lib/bitsandbytes/nn/modules.py -> build/bdist.linux-x86_64/egg/bitsandbytes/nn
copying build/lib/bitsandbytes/nn/triton_based_modules.py -> build/bdist.linux-x86_64/egg/bitsandbytes/nn
creating build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/adagrad.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/adam.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/adamw.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/lamb.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/lars.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/lion.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/optimizer.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/rmsprop.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
copying build/lib/bitsandbytes/optim/sgd.py -> build/bdist.linux-x86_64/egg/bitsandbytes/optim
creating build/bdist.linux-x86_64/egg/bitsandbytes/research
copying build/lib/bitsandbytes/research/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/research
creating build/bdist.linux-x86_64/egg/bitsandbytes/research/autograd
copying build/lib/bitsandbytes/research/autograd/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/research/autograd
copying build/lib/bitsandbytes/research/autograd/_functions.py -> build/bdist.linux-x86_64/egg/bitsandbytes/research/autograd
creating build/bdist.linux-x86_64/egg/bitsandbytes/research/nn
copying build/lib/bitsandbytes/research/nn/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/research/nn
copying build/lib/bitsandbytes/research/nn/modules.py -> build/bdist.linux-x86_64/egg/bitsandbytes/research/nn
creating build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/__init__.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/dequantize_rowwise.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/int8_matmul_mixed_dequanitze.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/int8_matmul_rowwise_dequantize.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/quantize_columnwise_and_transpose.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/quantize_global.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/quantize_rowwise.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/triton/triton_utils.py -> build/bdist.linux-x86_64/egg/bitsandbytes/triton
copying build/lib/bitsandbytes/libbitsandbytes_cuda118.so -> build/bdist.linux-x86_64/egg/bitsandbytes
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/__main__.py to __main__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/cextension.py to cextension.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/functional.py to functional.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/utils.py to utils.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/autograd/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/autograd/_functions.py to _functions.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup/env_vars.py to env_vars.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/cuda_setup/main.py to main.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/nn/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/nn/modules.py to modules.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/nn/triton_based_modules.py to triton_based_modules.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/adagrad.py to adagrad.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/adam.py to adam.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/adamw.py to adamw.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/lamb.py to lamb.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/lars.py to lars.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/lion.py to lion.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/optimizer.py to optimizer.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/rmsprop.py to rmsprop.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/optim/sgd.py to sgd.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/research/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/research/autograd/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/research/autograd/_functions.py to _functions.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/research/nn/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/research/nn/modules.py to modules.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/__init__.py to __init__.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/dequantize_rowwise.py to dequantize_rowwise.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/int8_matmul_mixed_dequanitze.py to int8_matmul_mixed_dequanitze.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/int8_matmul_rowwise_dequantize.py to int8_matmul_rowwise_dequantize.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/quantize_columnwise_and_transpose.py to quantize_columnwise_and_transpose.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/quantize_global.py to quantize_global.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/quantize_rowwise.py to quantize_rowwise.cpython-39.pyc
byte-compiling build/bdist.linux-x86_64/egg/bitsandbytes/triton/triton_utils.py to triton_utils.cpython-39.pyc
creating build/bdist.linux-x86_64/egg/EGG-INFO
copying bitsandbytes.egg-info/PKG-INFO -> build/bdist.linux-x86_64/egg/EGG-INFO
copying bitsandbytes.egg-info/SOURCES.txt -> build/bdist.linux-x86_64/egg/EGG-INFO
copying bitsandbytes.egg-info/dependency_links.txt -> build/bdist.linux-x86_64/egg/EGG-INFO
copying bitsandbytes.egg-info/top_level.txt -> build/bdist.linux-x86_64/egg/EGG-INFO
writing build/bdist.linux-x86_64/egg/EGG-INFO/native_libs.txt
zip_safe flag not set; analyzing archive contents...
bitsandbytes.cuda_setup.__pycache__.main.cpython-39: module references __file__
creating dist
creating 'dist/bitsandbytes-0.39.0-py3.9.egg' and adding 'build/bdist.linux-x86_64/egg' to it
removing 'build/bdist.linux-x86_64/egg' (and everything under it)
Processing bitsandbytes-0.39.0-py3.9.egg
removing '/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg' (and everything under it)
creating /root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg
Extracting bitsandbytes-0.39.0-py3.9.egg to /root/miniconda3/envs/py3.9/lib/python3.9/site-packages
bitsandbytes 0.39.0 is already the active version in easy-install.pth

Installed /root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg
Processing dependencies for bitsandbytes==0.39.0
Finished processing dependencies for bitsandbytes==0.39.0

===================================BUG REPORT===================================
Welcome to bitsandbytes. For bug reports, please run

python -m bitsandbytes

 and submit this information together with your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
================================================================================
bin /workspace/bitsandbytes/bitsandbytes/libbitsandbytes_cuda118.so
/workspace/bitsandbytes/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/nvidia/lib64'), PosixPath('/usr/local/nvidia/lib')}
  warn(msg)
/workspace/bitsandbytes/bitsandbytes/cuda_setup/main.py:149: UserWarning: /usr/local/nvidia/lib:/usr/local/nvidia/lib64 did not contain ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] as expected! Searching further paths...
  warn(msg)
/workspace/bitsandbytes/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('module'), PosixPath('//matplotlib_inline.backend_inline')}
  warn(msg)
/workspace/bitsandbytes/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIK1tFOFrWbmoa2ckCJYhzgBHKTSMeR/AeuScCCzugqlI utensilcandel@gmail.com')}
  warn(msg)
CUDA_SETUP: WARNING! libcudart.so not found in any environmental path. Searching in backup paths...
/workspace/bitsandbytes/bitsandbytes/cuda_setup/main.py:149: UserWarning: Found duplicate ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] files: {PosixPath('/usr/local/cuda/lib64/libcudart.so.11.0'), PosixPath('/usr/local/cuda/lib64/libcudart.so')}.. We'll flip a coin and try one of these, in order to fail forward.
Either way, this might cause trouble in the future:
If you get `CUDA error: invalid device function` errors, the above might be the cause and the solution is to make sure only one ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] in the paths that we search based on your env.
  warn(msg)
CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so.11.0
CUDA SETUP: Highest compute capability among GPUs detected: 8.6
CUDA SETUP: Detected CUDA version 118
CUDA SETUP: Loading binary /workspace/bitsandbytes/bitsandbytes/libbitsandbytes_cuda118.so...
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++ BUG REPORT INFORMATION ++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

++++++++++++++++++ /usr/local CUDA PATHS +++++++++++++++++++
/usr/local/cuda-11.8/compat/libcuda.so
/usr/local/cuda-11.8/targets/x86_64-linux/lib/libcudart.so
/usr/local/cuda-11.8/targets/x86_64-linux/lib/stubs/libcuda.so

+++++++++++++++ WORKING DIRECTORY CUDA PATHS +++++++++++++++
/workspace/bitsandbytes/bitsandbytes/libbitsandbytes_cuda118.so
/workspace/bitsandbytes/build/lib/bitsandbytes/libbitsandbytes_cuda118.so

++++++++++++++++++ LD_LIBRARY CUDA PATHS +++++++++++++++++++

++++++++++++++++++++++++++ OTHER +++++++++++++++++++++++++++
COMPILED_WITH_CUDA = True
COMPUTE_CAPABILITIES_PER_GPU = ['8.6']
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++
++++++++++++++++++++++ DEBUG INFO END ++++++++++++++++++++++
++++++++++++++++++++++++++++++++++++++++++++++++++++++++++++

Running a quick check that:
    + library is importable
    + CUDA function is callable


WARNING: Please be sure to sanitize sensible info from any such env vars!

SUCCESS!
Installation was successful!

In [11]:

%cd /workspace/axolotl

/workspace/axolotl

In [15]:

import wandb
wandb.login()

Failed to detect the name of this notebook, you can set it manually with the WANDB_NOTEBOOK_NAME environment variable to enable code saving.
wandb: Logging into wandb.ai. (Learn how to deploy a W&B server locally: https://wandb.me/wandb-server)
wandb: You can find your API key in your browser here: https://wandb.ai/authorize
wandb: Paste an API key from your profile and hit enter, or press ctrl+c to quit:

wandb: Appending key for api.wandb.ai to your netrc file: /root/.netrc

Out[15]:

True

In [16]:

!accelerate launch scripts/finetune.py examples/falcon/config-1b-lora.yml

Setting ds_accelerator to cuda (auto detect)

===================================BUG REPORT===================================
Welcome to bitsandbytes. For bug reports, please run

python -m bitsandbytes

 and submit this information together with your error trace to: https://github.com/TimDettmers/bitsandbytes/issues
================================================================================
bin /root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/libbitsandbytes_cuda118.so
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('/usr/local/nvidia/lib'), PosixPath('/usr/local/nvidia/lib64')}
  warn(msg)
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: /usr/local/nvidia/lib:/usr/local/nvidia/lib64 did not contain ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] as expected! Searching further paths...
  warn(msg)
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('module'), PosixPath('//matplotlib_inline.backend_inline')}
  warn(msg)
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: WARNING: The following directories listed in your path were found to be non-existent: {PosixPath('ssh-ed25519 AAAAC3NzaC1lZDI1NTE5AAAAIK1tFOFrWbmoa2ckCJYhzgBHKTSMeR/AeuScCCzugqlI utensilcandel@gmail.com')}
  warn(msg)
CUDA_SETUP: WARNING! libcudart.so not found in any environmental path. Searching in backup paths...
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/cuda_setup/main.py:149: UserWarning: Found duplicate ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] files: {PosixPath('/usr/local/cuda/lib64/libcudart.so.11.0'), PosixPath('/usr/local/cuda/lib64/libcudart.so')}.. We'll flip a coin and try one of these, in order to fail forward.
Either way, this might cause trouble in the future:
If you get `CUDA error: invalid device function` errors, the above might be the cause and the solution is to make sure only one ['libcudart.so', 'libcudart.so.11.0', 'libcudart.so.12.0'] in the paths that we search based on your env.
  warn(msg)
CUDA SETUP: CUDA runtime path found: /usr/local/cuda/lib64/libcudart.so.11.0
CUDA SETUP: Highest compute capability among GPUs detected: 8.6
CUDA SETUP: Detected CUDA version 118
CUDA SETUP: Loading binary /root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/libbitsandbytes_cuda118.so...
Setting ds_accelerator to cuda (auto detect)
WARNING:root:`trust_remote_code` is set to true. Please make sure that you reviewed the remote code/model.
INFO:root:loading tokenizer...
Using pad_token, but it is not set yet.
INFO:root:Loading prepared dataset from disk at last_run_prepared/546f844533b11547ae11d2027d0fd307...
INFO:root:Prepared dataset loaded from disk...
INFO:root:loading model and peft_config...
INFO:root:converting PEFT model w/ prepare_model_for_int8_training
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/peft/utils/other.py:76: FutureWarning: prepare_model_for_int8_training is deprecated and will be removed in a future version. Use prepare_model_for_kbit_training instead.
  warnings.warn(
INFO:root:found linear modules: ['query_key_value', 'dense_h_to_4h', 'dense_4h_to_h', 'dense']
trainable params: 12582912 || all params: 1324142592 || trainable%: 0.9502686550543342
INFO:root:Compiling torch model
INFO:root:Pre-saving adapter config to ./falcon-rw-1b
INFO:root:Starting trainer...
wandb: Currently logged in as: utensil. Use `wandb login --relogin` to force relogin
wandb: Tracking run with wandb version 0.15.3
wandb: Run data is saved locally in /workspace/axolotl/wandb/run-20230529_100931-jgmqihyz
wandb: Run `wandb offline` to turn off syncing.
wandb: Syncing run volcanic-paper-1
wandb: ⭐️ View project at https://wandb.ai/utensil/falcon-rw-1b
wandb: 🚀 View run at https://wandb.ai/utensil/falcon-rw-1b/runs/jgmqihyz
  0%|                                                | 0/108044 [00:00<?, ?it/s]You're using a GPT2TokenizerFast tokenizer. Please note that with a fast tokenizer, using the `__call__` method is faster than using a method to encode the text followed by a call to the `pad` method to get a padded encoding.
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/autograd/_functions.py:318: UserWarning: MatMul8bitLt: inputs will be cast from torch.float32 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/autograd/_functions.py:318: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")
{'loss': 1.5935, 'learning_rate': 7.5e-07, 'epoch': 0.0}                        
{'loss': 1.553, 'learning_rate': 1.5e-06, 'epoch': 0.0}                         
{'loss': 1.6538, 'learning_rate': 2.25e-06, 'epoch': 0.0}                       
{'loss': 2.0822, 'learning_rate': 3e-06, 'epoch': 0.0}                          
{'loss': 1.99, 'learning_rate': 3.75e-06, 'epoch': 0.0}                         
  0%|                                     | 5/108044 [00:03<19:20:06,  1.55it/s]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.17it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:50<00:00, 10.36it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.55it/s]
                                                                                
{'eval_loss': 2.321998119354248, 'eval_runtime': 51.5422, 'eval_samples_per_second': 10.593, 'eval_steps_per_second': 10.593, 'epoch': 0.0}
  0%|                                     | 5/108044 [00:55<19:20:06,  1.55it/s]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.34it/s]
{'loss': 3.1053, 'learning_rate': 4.5e-06, 'epoch': 0.0}                        
{'loss': 1.9726, 'learning_rate': 5.25e-06, 'epoch': 0.0}                       
{'loss': 1.8361, 'learning_rate': 6e-06, 'epoch': 0.0}                          
{'loss': 2.0174, 'learning_rate': 6.750000000000001e-06, 'epoch': 0.0}          
{'loss': 3.006, 'learning_rate': 7.5e-06, 'epoch': 0.0}                         
  0%|                                   | 10/108044 [00:58<132:16:11,  4.41s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.13it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:50<00:00, 10.31it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.45it/s]
                                                                                
{'eval_loss': 2.2817113399505615, 'eval_runtime': 51.5394, 'eval_samples_per_second': 10.594, 'eval_steps_per_second': 10.594, 'epoch': 0.0}
  0%|                                   | 10/108044 [01:49<132:16:11,  4.41s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.26it/s]
{'loss': 1.4985, 'learning_rate': 8.25e-06, 'epoch': 0.0}                       
{'loss': 1.3356, 'learning_rate': 9e-06, 'epoch': 0.0}                          
{'loss': 2.7513, 'learning_rate': 9.75e-06, 'epoch': 0.0}                       
{'loss': 0.9316, 'learning_rate': 1.05e-05, 'epoch': 0.0}                       
{'loss': 2.7089, 'learning_rate': 1.125e-05, 'epoch': 0.0}                      
  0%|                                   | 15/108044 [01:52<148:01:47,  4.93s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.11it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.29it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.46it/s]
                                                                                
{'eval_loss': 2.1745543479919434, 'eval_runtime': 51.6313, 'eval_samples_per_second': 10.575, 'eval_steps_per_second': 10.575, 'epoch': 0.0}
  0%|                                   | 15/108044 [02:44<148:01:47,  4.93s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.26it/s]
{'loss': 1.4429, 'learning_rate': 1.2e-05, 'epoch': 0.0}                        
{'loss': 1.5204, 'learning_rate': 1.275e-05, 'epoch': 0.0}                      
{'loss': 1.9816, 'learning_rate': 1.3500000000000001e-05, 'epoch': 0.0}         
{'loss': 2.4326, 'learning_rate': 1.4249999999999999e-05, 'epoch': 0.0}         
{'loss': 2.2008, 'learning_rate': 1.5e-05, 'epoch': 0.0}                        
  0%|                                   | 20/108044 [02:47<151:26:29,  5.05s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.10it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:50<00:00, 10.29it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:50<00:00, 10.48it/s]
                                                                                
{'eval_loss': 2.001559257507324, 'eval_runtime': 51.388, 'eval_samples_per_second': 10.625, 'eval_steps_per_second': 10.625, 'epoch': 0.0}
  0%|                                   | 20/108044 [03:38<151:26:29,  5.05s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.28it/s]
{'loss': 1.8915, 'learning_rate': 1.575e-05, 'epoch': 0.0}                      
{'loss': 1.7657, 'learning_rate': 1.65e-05, 'epoch': 0.0}                       
{'loss': 1.7492, 'learning_rate': 1.725e-05, 'epoch': 0.0}                      
{'loss': 1.7519, 'learning_rate': 1.8e-05, 'epoch': 0.0}                        
{'loss': 1.2162, 'learning_rate': 1.8750000000000002e-05, 'epoch': 0.0}         
  0%|                                   | 25/108044 [03:41<150:38:26,  5.02s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00,  9.99it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.19it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.36it/s]
                                                                                
{'eval_loss': 1.8274046182632446, 'eval_runtime': 51.5929, 'eval_samples_per_second': 10.583, 'eval_steps_per_second': 10.583, 'epoch': 0.0}
  0%|                                   | 25/108044 [04:32<150:38:26,  5.02s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.18it/s]
{'loss': 2.3359, 'learning_rate': 1.95e-05, 'epoch': 0.0}                       
{'loss': 1.6065, 'learning_rate': 2.025e-05, 'epoch': 0.0}                      
{'loss': 1.4764, 'learning_rate': 2.1e-05, 'epoch': 0.0}                        
{'loss': 1.6036, 'learning_rate': 2.175e-05, 'epoch': 0.0}                      
{'loss': 1.7938, 'learning_rate': 2.25e-05, 'epoch': 0.0}                       
  0%|                                   | 30/108044 [04:35<151:55:04,  5.06s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.14it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.28it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.32it/s]
                                                                                
{'eval_loss': 1.6766395568847656, 'eval_runtime': 51.7278, 'eval_samples_per_second': 10.555, 'eval_steps_per_second': 10.555, 'epoch': 0.0}
  0%|                                   | 30/108044 [05:27<151:55:04,  5.06s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.18it/s]
{'loss': 3.1058, 'learning_rate': 2.3250000000000003e-05, 'epoch': 0.0}         
{'loss': 1.7693, 'learning_rate': 2.4e-05, 'epoch': 0.0}                        
{'loss': 1.5462, 'learning_rate': 2.475e-05, 'epoch': 0.0}                      
{'loss': 1.2085, 'learning_rate': 2.55e-05, 'epoch': 0.0}                       
{'loss': 1.486, 'learning_rate': 2.625e-05, 'epoch': 0.0}                       
  0%|                                   | 35/108044 [05:30<151:45:53,  5.06s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.15it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.31it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.49it/s]
                                                                                
{'eval_loss': 1.6463074684143066, 'eval_runtime': 51.5896, 'eval_samples_per_second': 10.584, 'eval_steps_per_second': 10.584, 'epoch': 0.0}
  0%|                                   | 35/108044 [06:22<151:45:53,  5.06s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.31it/s]
{'loss': 1.6562, 'learning_rate': 2.7000000000000002e-05, 'epoch': 0.0}         
{'loss': 1.3435, 'learning_rate': 2.7750000000000004e-05, 'epoch': 0.0}         
{'loss': 0.9289, 'learning_rate': 2.8499999999999998e-05, 'epoch': 0.0}         
{'loss': 2.0553, 'learning_rate': 2.925e-05, 'epoch': 0.0}                      
{'loss': 1.7619, 'learning_rate': 3e-05, 'epoch': 0.0}                          
  0%|                                   | 40/108044 [06:25<151:23:26,  5.05s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.10it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:50<00:00, 10.28it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.46it/s]
                                                                                
{'eval_loss': 1.700196623802185, 'eval_runtime': 51.511, 'eval_samples_per_second': 10.6, 'eval_steps_per_second': 10.6, 'epoch': 0.0}
  0%|                                   | 40/108044 [07:16<151:23:26,  5.05s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.28it/s]
{'loss': 3.495, 'learning_rate': 2.9999999993654277e-05, 'epoch': 0.0}          
{'loss': 1.4613, 'learning_rate': 2.99999999746171e-05, 'epoch': 0.0}           
{'loss': 1.7003, 'learning_rate': 2.9999999942888465e-05, 'epoch': 0.0}         
  0%|                                   | 43/108044 [07:18<290:40:24,  9.69s/it]/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/autograd/_functions.py:318: UserWarning: MatMul8bitLt: inputs will be cast from torch.float32 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/autograd/_functions.py:318: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")
{'loss': 1.7253, 'learning_rate': 2.9999999898468384e-05, 'epoch': 0.0}         
{'loss': 1.6465, 'learning_rate': 2.999999984135685e-05, 'epoch': 0.0}          
  0%|                                   | 45/108044 [07:20<160:05:21,  5.34s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.15it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:50<00:00, 10.33it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.46it/s]
                                                                                
{'eval_loss': 1.6908812522888184, 'eval_runtime': 51.4094, 'eval_samples_per_second': 10.621, 'eval_steps_per_second': 10.621, 'epoch': 0.0}
  0%|                                   | 45/108044 [08:12<160:05:21,  5.34s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.30it/s]
{'loss': 1.5438, 'learning_rate': 2.999999977155386e-05, 'epoch': 0.0}          
{'loss': 0.7339, 'learning_rate': 2.999999968905942e-05, 'epoch': 0.0}          
{'loss': 1.7751, 'learning_rate': 2.9999999593873527e-05, 'epoch': 0.0}         
{'loss': 1.4588, 'learning_rate': 2.999999948599618e-05, 'epoch': 0.0}          
{'loss': 1.026, 'learning_rate': 2.9999999365427386e-05, 'epoch': 0.0}          
  0%|                                   | 50/108044 [08:15<151:52:36,  5.06s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:50<00:00, 10.21it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.38it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.53it/s]
                                                                                
{'eval_loss': 1.6710035800933838, 'eval_runtime': 51.6257, 'eval_samples_per_second': 10.576, 'eval_steps_per_second': 10.576, 'epoch': 0.0}
  0%|                                   | 50/108044 [09:06<151:52:36,  5.06s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.35it/s]
{'loss': 2.0806, 'learning_rate': 2.9999999232167138e-05, 'epoch': 0.0}         
{'loss': 1.3979, 'learning_rate': 2.999999908621544e-05, 'epoch': 0.0}          
{'loss': 1.6473, 'learning_rate': 2.9999998927572288e-05, 'epoch': 0.0}         
{'loss': 0.969, 'learning_rate': 2.999999875623769e-05, 'epoch': 0.0}           
{'loss': 1.8951, 'learning_rate': 2.9999998572211632e-05, 'epoch': 0.0}         
  0%|                                   | 55/108044 [09:09<151:30:18,  5.05s/it]
 99%|████████████████████████████████████████▍| 538/546 [00:50<00:00, 10.08it/s]
 99%|████████████████████████████████████████▌| 540/546 [00:51<00:00, 10.23it/s]
 99%|████████████████████████████████████████▋| 542/546 [00:51<00:00, 10.42it/s]
100%|████████████████████████████████████████▊| 544/546 [00:51<00:00, 10.53it/s]
                                                                                
{'eval_loss': 1.5737241506576538, 'eval_runtime': 51.8511, 'eval_samples_per_second': 10.53, 'eval_steps_per_second': 10.53, 'epoch': 0.0}
  0%|                                   | 55/108044 [10:01<151:30:18,  5.05s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.34it/s]
{'loss': 1.4765, 'learning_rate': 2.999999837549413e-05, 'epoch': 0.0}          
{'loss': 1.2403, 'learning_rate': 2.9999998166085172e-05, 'epoch': 0.0}         
{'loss': 1.3915, 'learning_rate': 2.999999794398476e-05, 'epoch': 0.0}          
{'loss': 1.0798, 'learning_rate': 2.9999997709192907e-05, 'epoch': 0.0}         
{'loss': 1.7814, 'learning_rate': 2.99999974617096e-05, 'epoch': 0.0}           
  0%|                                   | 60/108044 [10:04<151:49:38,  5.06s/it]
 99%|████████████████████████████████████████▍| 538/546 [00:50<00:00, 10.16it/s]
 99%|████████████████████████████████████████▌| 540/546 [00:51<00:00, 10.32it/s]
 99%|████████████████████████████████████████▋| 542/546 [00:51<00:00, 10.50it/s]
100%|████████████████████████████████████████▊| 544/546 [00:51<00:00, 10.60it/s]
                                                                                
{'eval_loss': 1.565449595451355, 'eval_runtime': 51.8417, 'eval_samples_per_second': 10.532, 'eval_steps_per_second': 10.532, 'epoch': 0.0}
  0%|                                   | 60/108044 [10:56<151:49:38,  5.06s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.40it/s]
{'loss': 1.6168, 'learning_rate': 2.999999720153484e-05, 'epoch': 0.0}          
{'loss': 1.4249, 'learning_rate': 2.999999692866863e-05, 'epoch': 0.0}          
{'loss': 1.6104, 'learning_rate': 2.9999996643110975e-05, 'epoch': 0.0}         
{'loss': 1.0135, 'learning_rate': 2.9999996344861868e-05, 'epoch': 0.0}         
{'loss': 0.7929, 'learning_rate': 2.9999996033921308e-05, 'epoch': 0.0}         
  0%|                                   | 65/108044 [10:59<151:41:37,  5.06s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:51<00:00, 10.15it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.32it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.48it/s]
                                                                                
{'eval_loss': 1.5621984004974365, 'eval_runtime': 51.9769, 'eval_samples_per_second': 10.505, 'eval_steps_per_second': 10.505, 'epoch': 0.0}
  0%|                                   | 65/108044 [11:51<151:41:37,  5.06s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.31it/s]
{'loss': 1.7342, 'learning_rate': 2.9999995710289307e-05, 'epoch': 0.0}         
{'loss': 1.7483, 'learning_rate': 2.9999995373965853e-05, 'epoch': 0.0}         
{'loss': 1.6172, 'learning_rate': 2.9999995024950948e-05, 'epoch': 0.0}         
{'loss': 1.4468, 'learning_rate': 2.9999994663244597e-05, 'epoch': 0.0}         
{'loss': 1.4673, 'learning_rate': 2.9999994288846798e-05, 'epoch': 0.0}         
  0%|                                   | 70/108044 [11:54<153:33:24,  5.12s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:51<00:00, 10.04it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.23it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.44it/s]
                                                                                
{'eval_loss': 1.5617916584014893, 'eval_runtime': 51.8986, 'eval_samples_per_second': 10.521, 'eval_steps_per_second': 10.521, 'epoch': 0.0}
  0%|                                   | 70/108044 [12:46<153:33:24,  5.12s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.29it/s]
{'loss': 1.5932, 'learning_rate': 2.999999390175755e-05, 'epoch': 0.0}          
{'loss': 1.2295, 'learning_rate': 2.999999350197686e-05, 'epoch': 0.0}          
{'loss': 1.6789, 'learning_rate': 2.999999308950472e-05, 'epoch': 0.0}          
{'loss': 1.3356, 'learning_rate': 2.999999266434113e-05, 'epoch': 0.0}          
{'loss': 2.3079, 'learning_rate': 2.99999922264861e-05, 'epoch': 0.0}           
  0%|                                   | 75/108044 [12:48<152:10:13,  5.07s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:51<00:00,  9.97it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.15it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.31it/s]
                                                                                
{'eval_loss': 1.5595687627792358, 'eval_runtime': 52.0458, 'eval_samples_per_second': 10.491, 'eval_steps_per_second': 10.491, 'epoch': 0.0}
  0%|                                   | 75/108044 [13:40<152:10:13,  5.07s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.13it/s]
{'loss': 2.4792, 'learning_rate': 2.999999177593962e-05, 'epoch': 0.0}          
{'loss': 0.88, 'learning_rate': 2.9999991312701694e-05, 'epoch': 0.0}           
{'loss': 1.6475, 'learning_rate': 2.9999990836772326e-05, 'epoch': 0.0}         
{'loss': 1.3836, 'learning_rate': 2.9999990348151514e-05, 'epoch': 0.0}         
{'loss': 1.7166, 'learning_rate': 2.9999989846839255e-05, 'epoch': 0.0}         
  0%|                                   | 80/108044 [13:43<152:10:19,  5.07s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:51<00:00,  9.92it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.10it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.28it/s]
                                                                                
{'eval_loss': 1.5077866315841675, 'eval_runtime': 52.3715, 'eval_samples_per_second': 10.426, 'eval_steps_per_second': 10.426, 'epoch': 0.0}
  0%|                                   | 80/108044 [14:36<152:10:19,  5.07s/it]
100%|█████████████████████████████████████████| 546/546 [00:52<00:00, 10.12it/s]
{'loss': 1.0969, 'learning_rate': 2.999998933283555e-05, 'epoch': 0.0}          
{'loss': 1.198, 'learning_rate': 2.9999988806140407e-05, 'epoch': 0.0}          
{'loss': 1.1006, 'learning_rate': 2.999998826675382e-05, 'epoch': 0.0}          
{'loss': 2.0751, 'learning_rate': 2.999998771467579e-05, 'epoch': 0.0}          
{'loss': 1.7348, 'learning_rate': 2.9999987149906314e-05, 'epoch': 0.0}         
  0%|                                   | 85/108044 [14:39<152:44:25,  5.09s/it]
 99%|████████████████████████████████████████▍| 539/546 [00:51<00:00, 10.13it/s]
 99%|████████████████████████████████████████▌| 541/546 [00:51<00:00, 10.28it/s]
 99%|████████████████████████████████████████▊| 543/546 [00:51<00:00, 10.49it/s]
                                                                                
{'eval_loss': 1.4808498620986938, 'eval_runtime': 51.9788, 'eval_samples_per_second': 10.504, 'eval_steps_per_second': 10.504, 'epoch': 0.0}
  0%|                                   | 85/108044 [15:31<152:44:25,  5.09s/it]
100%|█████████████████████████████████████████| 546/546 [00:51<00:00, 10.28it/s]
{'loss': 1.8124, 'learning_rate': 2.99999865724454e-05, 'epoch': 0.0}           
  0%|                                   | 86/108044 [15:31<579:50:48, 19.34s/it]/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/autograd/_functions.py:318: UserWarning: MatMul8bitLt: inputs will be cast from torch.float32 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")
/root/miniconda3/envs/py3.9/lib/python3.9/site-packages/bitsandbytes-0.39.0-py3.9.egg/bitsandbytes/autograd/_functions.py:318: UserWarning: MatMul8bitLt: inputs will be cast from torch.bfloat16 to float16 during quantization
  warnings.warn(f"MatMul8bitLt: inputs will be cast from {A.dtype} to float16 during quantization")
{'loss': 1.1838, 'learning_rate': 2.9999985982293046e-05, 'epoch': 0.0}         
{'loss': 1.3152, 'learning_rate': 2.9999985379449252e-05, 'epoch': 0.0}         
{'loss': 1.3686, 'learning_rate': 2.9999984763914014e-05, 'epoch': 0.0}         
{'loss': 1.3372, 'learning_rate': 2.9999984135687336e-05, 'epoch': 0.0}         
  0%|                                   | 90/108044 [15:35<156:43:27,  5.23s/it]
  0%|                                   | 90/108044 [15:45<314:54:55, 10.50s/it]

In [ ]: