Changes between Version 1 and Version 2 of kinaraai


Ignore:
Timestamp:
01/29/2026 06:03:56 PM (8 hours ago)
Author:
Ryan Erbstoesser
Comment:

update with new info.

Legend:

Unmodified
Added
Removed
Modified

  • kinaraai

    v1 v2  
    9090}}}
    9191
    92 == SDK
    93 
    94 Download the SDK from the Kinara support site.
    95 
    96 The file will be something like dv-sdk...tar
    97 
    98 Extract this file.
    99 
    100 
     92Program the Flash (one time only):
     93{{{
     94./bins/program_flash_aarch64 -e 0 -f ./boot_img/willow_therm.hex
     95<reboot the host/ reset ara device to get new firmware loaded on ara device>
     96}}}
     97
     98
     99Activate the chip:
     100{{{
     101root@noble-venice:~/art/linux/aarch64/hw_utils# ./bins/active_enable_aarch64 -e 0 -m 2
     102Kinara Inc. Silicon utilities version 0.5.9
     103
     104Run directory: /root/art/linux/aarch64/hw_utils
     105./bins/active_enable_aarch64 -e 0 -m 2
     106
     107[I:20260129:17:16:09:595108] [] [kinara_main_678][Active_Enable] In run phase of Flash image
     108[I:20260129:17:16:09:595458] [] [kinara_main_678][Active_Enable] Device Found Done ara-physical-id= 0
     109[I:20260129:17:16:09:595487] [] [kinara_main_678][IODeviceManager] enumerate ara2 pci devices
     110[I:20260129:17:16:09:794621] [] [kinara_main_678][IODeviceManager] path= /sys/devices/platform/soc@0/33800000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/
     111[I:20260129:17:16:09:794732] [] [kinara_main_678][IODeviceManager] instance name: IODeviceManager, found an ara2 pci device at: /sys/devices/platform/soc@0/33800000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/
     112[I:20260129:17:16:09:794880] [] [kinara_main_678][pci_io_00:01.0] cache management 0
     113[I:20260129:17:16:09:794901] [] [kinara_main_678][pci_io_00:01.0] instance name: Ara2PcieIO_0, ara2 pci device folder path: /sys/devices/platform/soc@0/33800000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/
     114[I:20260129:17:16:09:795008] [] [kinara_main_678][pci_io_00:01.0] instance name: Ara2PcieIO_0, phy0=0x18300000 phy1=0x18100000 phy2=0x18200000
     115[I:20260129:17:16:09:798594] [] [kinara_main_678][pci_io_00:01.0] enabling pcie_ste_dma for device
     116[W:20260129:17:16:09:798675] [] [kinara_main_678][pci_io_00:01.0] instance name: Ara2PcieIO_0, uiodma addr is : 0xf0a00000
     117[I:20260129:17:16:09:798777] [] [kinara_main_678][pci_io_00:01.0] successfully enabled pcie_ste_dma for pcie device
     118[I:20260129:17:16:09:805173] [] [kinara_main_678][pci_io_00:01.0] instance name: Ara2PcieIO_0, vendor=0x1e58 product=0x2
     119[I:20260129:17:16:09:807783] [] [kinara_main_678][pci_io_00:01.0] Calling pcie tl clock programming, removing pcie clock from bypass
     120[I:20260129:17:16:09:808154] [] [kinara_main_678][Active_Enable] [CTEST]: SBP Ready for Runtime Command
     121
     122[I:20260129:17:16:09:808182] [] [kinara_main_678][Active_Enable] [CTEST]: HIF is SETUP
     123[I:20260129:17:16:09:808197] [] [kinara_main_678][Active_Enable] Entering State Full PRD 2
     124[I:20260129:17:16:09:808214] [] [kinara_main_678][Active_Enable] {} 4
     125[I:20260129:17:16:09:808251] [] [kinara_main_678][Active_Enable] [CTEST]: Pushed HIF req Type :   211 1 1
     126[I:20260129:17:16:09:808265] [] [kinara_main_678][Active_Enable] [CTEST]: Pushed HIF req occ :  32
     127[I:20260129:17:16:09:808629] [] [kinara_main_678][Active_Enable] {} 4
     128[I:20260129:17:16:09:808673] [] [kinara_main_678][Active_Enable] [CTEST]: HIF response 0
     129[I:20260129:17:16:09:808693] [] [kinara_main_678][Active_Enable] [CTEST]: CMD_TYPE   211
     130[I:20260129:17:16:09:808711] [] [kinara_main_678][Active_Enable] [CTEST]: CMD_ID     1
     131[I:20260129:17:16:09:808728] [] [kinara_main_678][Active_Enable] [CTEST]: QID_ID     1
     132[I:20260129:17:16:09:808746] [] [kinara_main_678][Active_Enable] [CTEST]: HIF PS Read Response
     133[I:20260129:17:16:09:808765] [] [kinara_main_678][Active_Enable] [CTEST]: Status :   0
     134[I:20260129:17:16:09:808782] [] [kinara_main_678][Active_Enable] [CTEST]: PS Response Data Status :   0
     135[I:20260129:17:16:09:808799] [] [kinara_main_678][Active_Enable] [CTEST]: PS Response Data Current Power State :   2
     136[I:20260129:17:16:09:808816] [] [kinara_main_678][Active_Enable] [CTEST]: PS Response Data PVT_LOCK :  0
     137[I:20260129:17:16:09:808836] [] [kinara_main_678][Active_Enable] [CTEST]: TEST PASSED
     138}}}
     139
     140Configure the DRAM:
     141{{{
     142root@noble-venice:~/art/linux/aarch64/hw_utils# ./bins/ddr_unified_config_aarch64 -e 0 -s 1 -g 1 -l 3 -w 2 -f 0 -n 0 -u 0 -m 0 -r 0 -b willow -o ./ddr_config/ddr_cfg_35.bin
     143Kinara Inc. Silicon utilities version 0.5.9
     144
     145Run directory: /root/art/linux/aarch64/hw_utils
     146./bins/ddr_unified_config_aarch64 -e 0 -s 1 -g 1 -l 3 -w 2 -f 0 -n 0 -u 0 -m 0 -r 0 -b willow -o ./ddr_config/ddr_cfg_35.bin
     147
     148[I:20260129:17:17:05:155740] [] [kinara_main_706][IODeviceManager] enumerate ara2 pci devices
     149[I:20260129:17:17:05:343281] [] [kinara_main_706][IODeviceManager] path= /sys/devices/platform/soc@0/33800000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/
     150[I:20260129:17:17:05:343392] [] [kinara_main_706][IODeviceManager] instance name: IODeviceManager, found an ara2 pci device at: /sys/devices/platform/soc@0/33800000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/
     151[I:20260129:17:17:05:343538] [] [kinara_main_706][pci_io_00:01.0] cache management 0
     152[I:20260129:17:17:05:343558] [] [kinara_main_706][pci_io_00:01.0] instance name: Ara2PcieIO_0, ara2 pci device folder path: /sys/devices/platform/soc@0/33800000.pcie/pci0000:00/0000:00:00.0/0000:01:00.0/
     153[I:20260129:17:17:05:343663] [] [kinara_main_706][pci_io_00:01.0] instance name: Ara2PcieIO_0, phy0=0x18300000 phy1=0x18100000 phy2=0x18200000
     154[I:20260129:17:17:05:346269] [] [kinara_main_706][pci_io_00:01.0] enabling pcie_ste_dma for device
     155[W:20260129:17:17:05:346327] [] [kinara_main_706][pci_io_00:01.0] instance name: Ara2PcieIO_0, uiodma addr is : 0xf0a00000
     156[I:20260129:17:17:05:346428] [] [kinara_main_706][pci_io_00:01.0] successfully enabled pcie_ste_dma for pcie device
     157[I:20260129:17:17:05:351159] [] [kinara_main_706][pci_io_00:01.0] instance name: Ara2PcieIO_0, vendor=0x1e58 product=0x2
     158[I:20260129:17:17:05:353537] [] [kinara_main_706][pci_io_00:01.0] Calling pcie tl clock programming, removing pcie clock from bypass
     159[I:20260129:17:17:05:353868] [] [kinara_main_706][ddr_unified_config] Configuring DDR now. Approxmiate time to complete: 60 seconds.
     160[I:20260129:17:17:05:353894] [] [kinara_main_706][ddr_unified_config] UNIFIED CFG BIN LOADING
     161[I:20260129:17:17:36:009795] [] [kinara_main_706][ddr_unified_config] DDR configuation successful. DDR BIST: PASSED
     162}}}
     163
     164
     165Activate Proxy: (change config/proxy_config.yaml if you want to run in background)
     166
     167{{{
     168root@noble-venice:~/art/linux/aarch64/proxy# sudo ./proxy_aarch64 -c ../../config/proxy_config.yaml
     169
     170Kinara Inc. Silicon utilities version 0.5.9
     171
     172Run directory: /root/art/linux/aarch64/proxy
     173./proxy_aarch64 -c ../../config/proxy_config.yaml
     174
     175[I:20260129:17:23:04:425472] [] [kinara_main_764][GlobalLogger]
     176root@noble-venice:~/art/linux/aarch64/proxy# ps -aef | grep prox
     177root         766       1  9 17:23 ?        00:00:00 ./proxy_aarch64 -c ../../config/proxy_config.yaml
     178}}}
     179
     180Download a sample yolov8n model (.dvm file) and untar:
     181 * Download [https://workdrive.zohoexternal.com/external/8c92f49dc6ba6e433a99801eef80363a46c7c8b74c9907bd47b0f3c4b9eccf14 URL] (may require access to support.kinara.ai portal)
     182 * Transfer to Gateworks SBC and untar:
     183{{{
     184root@noble-venice:~/art/linux/aarch64/nnapp# wget 172.24.20.26/kinara/sdk_model_compile.tar.gz
     185--2026-01-29 17:26:52--  http://192.168.1.1/kinara/sdk_model_compile.tar.gz
     186Connecting to 172.24.20.26:80... connected.
     187HTTP request sent, awaiting response... 200 OK
     188Length: 23754978 (23M) [application/x-gtar-compressed]
     189Saving to: ‘sdk_model_compile.tar.gz’
     190
     191sdk_model_compile.tar.gz                        100%[=====================================================================================================>]  22.65M   112MB/s    in 0.2s   
     192
     1932026-01-29 17:26:52 (112 MB/s) - ‘sdk_model_compile.tar.gz’ saved [23754978/23754978]
     194root@noble-venice:~/art/linux/aarch64/nnapp# tar -xvf sdk_model_compile.tar.gz
     195}}}
     196
     197Either change the input path to the model, or move the model to /tmp
     198
     199Modify the ../../config/nnapp_config.yaml and change the input_path in the async section:
     200{{{
     201async:
     202        # if this flag is TRUE, then the input directory of each model is iterated only once.
     203        # else, the input directory is iterated for model_param.batch_size * model_param.iterations times
     204        run_once_and_for_all: False
     205        utilization_stats: False
     206        # dump outputs with timestamp
     207        dump_with_timestamp: False
     208        # this section contains all the model related params for infer async mode
     209        models:
     210          0:
     211            name: "model1"
     212            path: "/tmp/model.dvm"
     213            input_path: ""
     214            output_path: ""
     215
     216}}}
     217
     218Execute the sample model run:
     219{{{
     220root@noble-venice:~/art/linux/aarch64/nnapp# ./nnapp_aarch64 --config ../../config/nnapp_config.yaml
     221[W:20260129:17:29:41:832406] [] [n/a_-1][config] default YAML file not found  ./config.yaml , specify yaml path
     222Kinara, Inc.
     223
     224*---------------------------------------------------------------*
     225*              Neural Network Inference Application             *
     226*    Copyright(c) 2018-23, Kinara, Inc. All rights reserved.    *
     227*---------------------------------------------------------------*
     228
     229Run:
     230working directory: /root/art/linux/aarch64/nnapp
     231./nnapp_aarch64 --config ../../config/nnapp_config.yaml
     232the app expects the inputs as raw tensors. The inputs are assumed to be preprocessed and written to a file
     233the inputs have to be in the below format:
     234<input_image_name>/quantized/layer<layer_id>-<blob_name>-<blob_id>_input_dv.dat
     235and the golden inputs in the below format:
     236<input_image_name>/quantized/<layer_name>-<layer_id>-<blob_name>-<blob_id>
     237[I:20260129:17:29:41:898866] [] [main_807][model_parameters] empty input_path for model= model1 , setting output_path and golden_output_path to none
     238[I:20260129:17:29:41:903483] [] [main_807][BaseApp] connected to proxy socket: IP_Address= 127.0.0.1  port= 5000
     239[I:20260129:17:29:41:904161] [] [main_807][BaseApp]
     240 +------------+-----------------+
     241| Product    | Current Version |
     242+------------+-----------------+
     243| client_lib | 1.3.0.0         |
     244| firmware   | 0.128.11.0      |
     245| proxy      | 1.3.0.0         |
     246| sysapi     | 1.1.61.0        |
     247+------------+-----------------+
     248
     249[I:20260129:17:29:41:904950] [] [main_807][BaseApp]
     250 +------------+--------------------+
     251| Product    | Supported Versions |
     252+------------+--------------------+
     253| client_lib | 1.0.0.0            |
     254| client_lib | 1.1.1.0            |
     255| client_lib | 1.1.2.0            |
     256| client_lib | 1.3.0.0            |
     257| cnn_model  | 2.0.0.0            |
     258| cnn_model  | 2.1.0.0            |
     259| firmware   | 0.5.2.0            |
     260| firmware   | 1.1.2.0            |
     261| llm_model  | 3.0.0.0            |
     262| llm_model  | 3.1.0.0            |
     263| pci_driver | 1.0.4.0            |
     264| pci_driver | 1.0.6.6            |
     265| proxy      | 0.8.0.0            |
     266| proxy      | 0.9.0.0            |
     267| proxy      | 1.1.1.0            |
     268| proxy      | 1.3.0.0            |
     269+------------+--------------------+
     270
     271[I:20260129:17:29:41:905463] [] [main_807][BaseApp] found endpoints: count= 1
     272[I:20260129:17:29:41:905908] [] [main_807][BaseApp] creation of endpoint group successful: model= model1
     273[I:20260129:17:29:41:984637] [] [main_807][BaseApp] model load successful: model= model1
     274[I:20260129:17:29:42:011215] [] [task_print_device_stats_812][Stats] temperature= 36.2557 ep status:  DV_ENDPOINT_STATE_ACTIVE  0x1000608c:   15d0200
     275[I:20260129:17:29:42:072253] [] [main_807][AsyncApp] completed iteration= 1
     276[I:20260129:17:29:42:145357] [] [main_807][AsyncApp] completed iteration= 2
     277[I:20260129:17:29:42:218729] [] [main_807][AsyncApp] completed iteration= 3
     278[I:20260129:17:29:42:291966] [] [main_807][AsyncApp] completed iteration= 4
     279[I:20260129:17:29:42:365203] [] [main_807][AsyncApp] completed iteration= 5
     280[I:20260129:17:29:42:437872] [] [main_807][AsyncApp] completed iteration= 6
     281[I:20260129:17:29:42:510537] [] [main_807][AsyncApp] completed iteration= 7
     282[I:20260129:17:29:42:584005] [] [main_807][AsyncApp] completed iteration= 8
     283[I:20260129:17:29:42:656563] [] [main_807][AsyncApp] completed iteration= 9
     284[I:20260129:17:29:42:729354] [] [main_807][AsyncApp] completed iteration= 10
     285[I:20260129:17:29:42:802121] [] [main_807][AsyncApp] completed iteration= 11
     286[I:20260129:17:29:42:874931] [] [main_807][AsyncApp] completed iteration= 12
     287[I:20260129:17:29:42:947354] [] [main_807][AsyncApp] completed iteration= 13
     288[I:20260129:17:29:43:012893] [] [task_print_device_stats_812][Stats] temperature= 36.2557 ep status:  DV_ENDPOINT_STATE_ACTIVE  0x1000608c:   15d0200
     289[I:20260129:17:29:43:020108] [] [main_807][AsyncApp] completed iteration= 14
     290[I:20260129:17:29:43:094043] [] [main_807][AsyncApp] completed iteration= 15
     291[I:20260129:17:29:43:166269] [] [main_807][AsyncApp] completed iteration= 16
     292[I:20260129:17:29:43:238760] [] [main_807][AsyncApp] completed iteration= 17
     293[I:20260129:17:29:43:311065] [] [main_807][AsyncApp] completed iteration= 18
     294[I:20260129:17:29:43:383388] [] [main_807][AsyncApp] completed iteration= 19
     295[I:20260129:17:29:43:455880] [] [main_807][AsyncApp] completed iteration= 20
     296[I:20260129:17:29:43:456773] [] [main_807][Stats] summary stats
     297 ╔════════╤════════════════════════════╤═════════════════════════╤═══════════════════════════╤══════════════════╗
     298║ Model  │        hw_time (ms)        │      ip_time (ms)       │       op_time (ms)        │  app_time (ms)   ║
     299╠════════╪════════╤═════════╤═════════╪═══════╤═══════╤═════════╪════════╤════════╤═════════╪═════╤═════╤══════╣
     300║        │ max    │ min     │ mean    │ max   │ min   │ mean    │ max    │ min    │ mean    │ max │ min │ mean ║
     301╟────────┼────────┼─────────┼─────────┼───────┼───────┼─────────┼────────┼────────┼─────────┼─────┼─────┼──────╢
     302║ model1 │ 3.1052 │ 3.08395 │ 3.09515 │ 4.365 │ 2.346 │ 3.17945 │ 12.187 │ 10.369 │ 11.3023 │ 76  │ 71  │ 71.9 ║
     303╚════════╧════════╧═════════╧═════════╧═══════╧═══════╧═════════╧════════╧════════╧═════════╧═════╧═════╧══════╝
     304 
     305  model_name=model1 iteration=20 batch_size=5 num_ip=1 num_op=1 total_ip_size=122880000 total_op_size=120960000 num_inferences=100
     306
     307[I:20260129:17:29:43:457176] [] [main_807][Stats] endpoint load distribution:  [0,100]
     308[I:20260129:17:29:43:457195] [] [main_807][Stats] total number of inferences:  100
     309[I:20260129:17:29:43:457221] [] [main_807][Stats] Performance: HW IPS= 323.086  App IPS= 69.541
     310root@noble-venice:~/art/linux/aarch64/nnapp#
     311}}}
     312
     313Note the final line that shows Performance: HW IPS= 323.086.
     314
     315