Supported Test Vector Configurations

Aerial SDK 22-4

This release of the cuBB SDK currently supports the following test-vector configurations.

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

7201 7201 base 1 1 1
7202 7203 mcsTable 2 2 2
7204 7204 mcs 1 1 1
7205 7207 num of layers 3 3 3
7208 7208 rb0, Nrb 1 1 1
7209 7210 sym0 2 2 2
7211 7211 dmsr0 1 1 1
7212 7213 Nsym 2 2 2
7214 7214 SCID 1 1 1
7215 7215 BWP0, nBWP 1 1 1
7216 7216 RNTI 1 1 1
7217 7219 addPos 3 3 3
7220 7220 dataScId 1 1 1
7221 7222 maxLen 2 2 2
7223 7223 dmrsScId 1 1 1
7224 7224 nCdm 1 1 1
7225 7225 port0 1 1 1
7226 7228 nAnt 3 2 1
7229 7229 slotIdx 1 1 1
7230 7232 rvIdx 3 3 3
7233 7235 FDM 3 3 3
7236 7241 CDM 6 6 5
7242 7244 rvIdx>0/BGN=1 3 3 3
7245 7245 ulGridSize=106 1 1 0
7246 7247 dmrs_par per Ueg 2 2 2
7248 7250 additional FDM 3 3 3
7251 7257 precoding 7 7 7
7258 7260 mapping type B 3 3 3
7261 7272 Flexible DMRS ports 12 12 11
7273 7273 MCS > 28 1 1 1
7274 7279 additional nCDM=1 6 6 6
7280 7283 Flexible SLIV 4 4 4
7301 7320 multi-params 20 20 18
7321 7323 LBRM 3 3 3
7324 7326 HARQ-rx 3 3 0
7327 7330 8/16 UEs 4 4 4
7331 7338 multiple layers 8 8 8
7340 7340 Multi-layers with nAnt=16 1 1 0
7401 7403 CFO 3 3 3
7404 7406 TO 3 3 3
7407 7407 RSSI 1 1 1
7408 7408 CFO w/ SDM 1 1 1
7409 7409 TO w/ SDM 1 1 1
7410 7411 CEE-TDI 2 2 2
7412 7413 rx power 2 2 2
7414 7414 TDI maxLen = 2 1 1 1
7415 7417 small/big/zero rx 3 3 3
7418 7419 additional TDI 2 2 2
7420 7426 IRC=0 7 7 7
7427 7432 SINR meas 6 6 6
7501 7516 UCI on PUSCH (w/o data) 16 16 16
7517 7530 UCI on PUSCH (w/ data) 14 14 14
7531 7531 UciOnPusch DTX 1 1 0
7532 7532 UciOnPusch CRC fail 1 1 0
7533 7534 UciOnPusch addPos 2 2 2
7551 7570 UciOnPusch (multi-params) 20 20 20
7571 7575 UCI w/ and w/o data 5 5 1
7601 7613 FR1 BW mu = 1 13 13 1
7614 7621 FR1 BW mu = 0 8 8 0
7901 7901 demo_msg3 1 1 1
7902 7902 demo_traffic_ul 1 1 1
7903 7904 UciOnPusch conformance 0 0 0
7016 7153 sweep Zc/mcs (skip 7016,7017,7024,7025,7032,7039,7045,7057) 130 130 128

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

6001 6003 bases for format 0 3 3 3
6004 6010 vary single parameter for format 0 7 7 7
6011 6040 vary multiple parameters for format 0 30 30 30
6041 6046 vary slotIdx (single-UCI) for format 0 6 6 6
6047 6056 multi-UCI tests for format 0 10 10 10
6057 6061 vary slotIdx (multi-UCI) for format 0 5 5 5
6101 6103 bases for format 1 3 3 3
6104 6116 vary single parameter for format 1 13 13 13
6117 6146 vary multiple parameters for format 1 30 30 30
6147 6155 vary slotIdx (single-UCI) for format 1 9 9 9
6156 6173 multi-UCI tests for format 1 18 18 18
vary slotIdx (multi-UCI) for format 1
6201 6203 bases for format 2 3 3 3
6204 6219 test Nf for format 2 16 16 16
6220 6235 test Nt and freq hopping for format 2 16 16 16
6236 6236 11 info bits and 2 PRBS for format 2 1 1 1
6239 6245 different payload sizes for format 2 7 7 7
6301 6310 bases for format 3 10 10 10
6311 6313 multi-UCI tests for format 3 3 3 3
6314 6324 tests with freqHop enabled for format 3 11 11 11
6325 6335 tests with freqHop disabled for format 3 11 11 11
6336 6346 tests with add’l DMRS postion, freqHop enabled for format 3 11 11 11
6347 6357 tests with add’l DMRS postion, freqHop disabled for format 3 11 11 11
6358 6364 different payload sizes for format 3 7 7 7
6365 6373 24-UCI tests for format 3 9 9 9
6501 6513 sweep different bandwidth for format 0, mu = 1 13 13 13
6514 6526 sweep different bandwidth for format 1, mu = 1 13 13 13
6527 6539 sweep different bandwidth for format 2, mu = 1 13 13 13
6540 6552 sweep different bandwidth for format 3, mu = 1 13 13 13
6553 6560 sweep different bandwidth for format 0, mu = 0 8 8 0
6561 6568 sweep different bandwidth for format 1, mu = 0 8 8 0
6569 6576 sweep different bandwidth for format 2, mu = 0 8 8 0
6577 6584 sweep different bandwidth for format 3, mu = 0 8 8 0
6585 6586 rx power for format 0 2 2 2
6587 6588 rx power for format 1 2 2 2
6589 6590 rx power for format 2 2 2 2
6591 6592 rx power for format 3 2 2 2
6593 6595 very small/very big/forcRxZero rx power for format 0 3 3 3
6596 6598 very small/very big/forcRxZero rx power for format 1 3 3 3
6599 6601 very small/very big/forcRxZero rx power for format 2 3 3 3
6602 6605 very small/very big/forcRxZero rx power for format 3 4 4 4
6801 6802 perf TV F08 2 2 2
6803 6804 perf TV F14 2 2 2

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

5001 5001 base 1 1 1
5002 5002 format 0 1 1 0
5003 5003 rootIdx 1 1 1
5004 5004 zoneIdx 1 1 1
5005 5005 prmbIdx 1 1 1
5006 5006 Nant 1 1 0
5007 5007 N_nc 1 1 1
5008 5008 delay 1 1 1
5009 5009 SNR 1 1 1
5010 5010 CFO 1 1 1
5011 5011 2-UE 1 1 1
5012 5012 4-UE 1 1 1
5013 5013 4FDM/16UE 1 1 1
5014 5018 rx power 5 5 5
5101 5101 FDD,mu=0,B4,nAnt=2 1 1 0
5102 5102 FDD,mu=1,B4,nAnt=4 1 1 1
5103 5103 TDD,mu=0,B4,nAnt=8 1 1 0
5104 5104 TDD,mu=1,B4,nAnt=16 1 1 0
5105 5105 FDD,mu=0,F0,nAnt=16 1 1 0
5106 5106 FDD,mu=1,F0,nAnt=8 1 1 0
5107 5107 TDD,mu=0,F0,nAnt=4 1 1 0
5108 5108 TDD,mu=1,F0,nAnt=2 1 1 0
5201 5213 FR1 BW mu = 1 13 13 13
5214 5221 FR1 BW mu = 0 8 8 0
5801 5802 perf TV F08 2 2 2
5803 5804 perf TV F14 2 2 0
5901 5901 demo_msg1 1 1 1
5911 5914 comformance TC 4 4 0

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

3201 3201 base 1 1 1
3202 3203 mcsTable 2 2 2
3204 3204 mcs 1 1 1
3205 3207 num of layers 3 3 3
3208 3208 rb0, Nrb 1 1 1
3209 3210 sym0 2 2 2
3211 3211 dmrs0 1 1 1
3212 3213 Nsym 2 2 2
3214 3214 SCID 1 1 0
3215 3215 BWP0, nBWP 1 1 1
3216 3216 RNTI 1 1 1
3217 3219 addPos 3 3 3
3220 3220 dataScId 1 1 1
3221 3222 maxLen 2 2 2
3223 3223 dmrsScId 1 1 1
3224 3224 nCdm 1 1 1
3225 3225 port0 1 1 0
3226 3228 nAnt 3 3 1
3229 3229 slotIdx 1 1 1
3230 3232 rvIdx 3 3 3
3233 3235 FDM 3 3 3
3236 3241 SDM/SCID 6 6 4
3242 3244 rvIdx>0/BGN=1 3 3 3
3245 3245 dlGridSize=106 1 1 0
3246 3247 dmrs_par per Ueg 2 2 1
3248 3254 precoding 7 7 4
3255 3257 mapping type B 3 3 3
3258 3260 mixed precoding 3 3 1
3261 3261 refPoint 1 1 1
3262 3262 TxPower 1 1 1
3263 3263 modComp
3264 3264 precoding (mixed nPorts) 1 1 0
3265 3265 TxPower with 2 UEs 1 1 1
3266 3267 different rv 2 2 2
3268 3269 multi-layer 2 2 0
3271 3276 nCDM = 1 6 6 5
3321 3322 LBRM 2 2 2
3323 3333 RE map from CSI-RS 11 11 11
3334 3336 8/16 UEs (SU-MIMO) 3 3 3
3337 3337 16 UEs (MU-MIMO) 1 1 0
3401 3413 FR1 BW mu = 1 13 13 1
3414 3421 FR1 BW mu = 0 8 8 0
3901 3901 demo_coreset0 1 1 1
3902 3902 demo_msg2 1 1 1
3903 3903 demo_msg4 1 1 1
3904 3904 demo_traffic_dl 1 1 1
3001 3015 multi-params 15 15 15
3016 3154 sweep Zc/mcs (3016,3017,3024,3025,3032,3039,3045,3057 are skipped) 131 131 131

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

2001 2001 base 1 1 0
2002 2002 slotIdx 1 1 1
2003 2003 nBWP 1 1 1
2004 2004 BPW0 1 1 1
2005 2005 sym0 1 1 1
2006 2007 Nsym 2 2 2
2008 2009 crstIdx 2 2 2
2010 2010 intl 1 1 1
2011 2012 nBndl 2 2 2
2013 2014 nIntl 2 2 2
2015 2015 nShift 1 1 1
2016 2016 isCSS 1 1 1
2017 2017 rnti 1 1 1
2018 2018 scrbId 1 1 1
2019 2019 scrbRnti 1 1 1
2020 2022 aggrL 3 3 3
2023 2023 dbQam 1 1 1
2024 2024 dbDmrs 1 1 1
2025 2025 Npayload 1 1 1
2026 2027 crstMap 2 2 2
2028 2028 nDCI 1 1 1
2029 2029 Npayload 1 1 1
2030 2030 aggrL 1 1 1
2031 2031 precoding 1 1 1
2032 2032 modComp
2033 2033 multi-PDCCH 1 1 1
2101 2112 multi-params 12 12 12
2201 2213 FR1 BW mu = 1 13 13 1
2214 2221 FR1 BW mu = 0 8 8 0
2801 2802 perf TV F14 2 2 2
2803 2804 perf TV F08 2 2 2
2805 2806 perf TV F09 2 2 2
2901 2901 demo_msg2 1 1 1
2902 2902 demo_msg4 1 1 1
2903 2903 demo_coreset0 1 1 1
2904 2904 demo_traffic_dl 1 1 1
2905 2905 demo_msg5 1 1 1

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

1001 1001 base 1 1 1
1002 1002 mu = 0 1 1 0
1003 1003 N_CELL_ID 1 1 1
1004 1004 n_hf = 1 1 1 1
1005 1005 L_max = 4 1 1 1
1006 1006 k_SSB 1 1 1
1007 1007 offsetPointA 1 1 1
1008 1008 SFN 1 1 1
1009 1009 blockIdx 1 1 1
1010 1010 precoding 1 1 1
1011 1011 betaPss 1 1 1
1101 1101 mu=0, 1SSB 1 1 0
1102 1102 mu=1, 1SSB 1 1 1
1103 1103 mu=1, 2SSB 1 1 0
1104 1104 mu=1, 2SSB 1 1 1
1202 1213 FR1 BW, mu = 1 12 12 7
1214 1221 FR1 BW, mu = 0 8 8 0
1801 1801 Perf TV 1 1 1
1901 1901 demo_ssb 1 1 1
1902 1902 for CP pipeline 1 1 1

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

4001 4004 nPorts = 1 4 4 4
4005 4007 nPorts = 2 3 3 1
4008 4009 nPorts = 4 2 2 2
4010 4038 nPorts > 4, row > 5 29 29 0
4039 4039 RB0 1 1 1
4040 4040 nRB 1 1 1
4041 4041 sym0 1 1 1
4042 4042 sym1 1 1 0
4043 4043 nID 1 1 1
4044 4044 power control 1 1 0
4045 4050 freqDomainAllocation 6 6 3
4051 4051 idxSlot 1 1 1
4052 4054 batching 3 3 3
4055 4055 small gird size 1 1 0
4056 4056 TRS 1 1 1
4057 4057 precoding
4058 4058 modComp
4059 4060 16/32 CSIRS PDUs 2 2 2
4101 4103 multiple parameters 3 3 0
4201 4213 FR1 BW mu = 1 13 13 0
4214 4221 FR1 BW mu = 0 8 8 0
4801 4801 perf TV F08 1 1 1
4802 4802 perf TV F09 1 1 0
4803 4803 perf TV F14 1 1 0

TC start

TC end

Description

TV generated

cuPHY pass

cuBB pass

90001 90007 single channel 7 7 7
90011 90012 dlmix/ulmix 2 2 2
90013 90015 s-slot 3 3 3
90016 90018 multi-cell base case 3 3 3
90019 90019 prcd+noPrcd 1 1 1
90020 90020 BFP14+BFP9 1 1 1
90021 90022 HARQ 2 2 2
90023 90023 empty slot 1 1 1

The ldpc\_perf\_collect.py Python script from the cuPHY repository can be used to perform error rate tests for the cuPHY LDPC decoder. Currently, there are test input files defined for Z = [64, 128, 256, 384], BG = [1,2]. The current tests check whether the block error rate (BLER, also sometimes referred to as Frame Error Rate or FER) is less than 0.1.

From the build directory, the following commands will run the tests:

Copy
Copied!
            

../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG1\_Z64\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG1\_Z128\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG1\_Z256\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG1\_Z384\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG2\_Z64\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG2\_Z128\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG2\_Z256\_BLER0.1.txt -f -w 800 -P ../util/ldpc/ldpc\_perf\_collect.py --mode test -i ../util/ldpc/test/ldpc\_decode\_BG2\_Z384\_BLER0.1.txt -f -w 800 -P

Each test input file contains multiple tests for different code rates, as specified by the number of parity nodes. The format of the input files has the following form:

Copy
Copied!
            

# BG Z num_parity num_iter SNR max_BER max_BLER #-------------------------------------------------------------- 1 384 4 10 6.87 1 0.1 1 384 5 10 6.15 1 0.1 1 384 6 10 5.64 1 0.1 1 384 7 10 5.17 1 0.1 1 384 8 10 4.79 1 0.1 ...

After running each of the test cases, the ldpc\_perf\_collect.py script will display an output table:

Copy
Copied!
            

+-----------------------------------------------------------------------------------------------------------+ | # BG Z num\_parity num\_iter SNR max\_BER BER max\_BLER BLER STATUS | | | | 1 384 4 10 6.870 1.000000e+00 4.833980e-04 1.000000e-01 8.750000e-02 PASS | | | | 1 384 5 10 6.150 1.000000e+00 1.481120e-04 1.000000e-01 7.250000e-02 PASS | | | | 1 384 6 10 5.640 1.000000e+00 5.652230e-05 1.000000e-01 8.000000e-02 PASS | | | | 1 384 7 10 5.170 1.000000e+00 7.886480e-05 1.000000e-01 8.750000e-02 PASS | | | | 1 384 8 10 4.790 1.000000e+00 1.673470e-04 1.000000e-01 8.375000e-02 PASS | | | | 1 384 9 10 4.480 1.000000e+00 1.185190e-04 1.000000e-01 7.625000e-02 PASS | | | | 1 384 10 10 4.200 1.000000e+00 8.552320e-05 1.000000e-01 8.875000e-02 PASS | | | | 1 384 11 10 3.920 1.000000e+00 5.385890e-05 1.000000e-01 8.375000e-02 PASS | | | | 1 384 12 10 3.660 1.000000e+00 1.234020e-04 1.000000e-01 9.125000e-02 PASS | | | | 1 384 13 10 3.450 1.000000e+00 7.013490e-05 1.000000e-01 8.000000e-02 PASS | | | | 1 384 14 10 3.220 1.000000e+00 7.620150e-05 1.000000e-01 8.125000e-02 PASS | | | | 1 384 15 10 3.020 1.000000e+00 5.800190e-05 1.000000e-01 7.250000e-02 PASS | | | | 1 384 16 10 2.830 1.000000e+00 8.774270e-05 1.000000e-01 8.375000e-02 PASS | | | | 1 384 17 10 2.640 1.000000e+00 4.838420e-05 1.000000e-01 7.750000e-02 PASS | | | | 1 384 18 10 2.500 1.000000e+00 3.950640e-05 1.000000e-01 7.875000e-02 PASS | | | | 1 384 19 10 2.310 1.000000e+00 3.551140e-05 1.000000e-01 8.375000e-02 PASS | | | | 1 384 20 10 2.150 1.000000e+00 2.500590e-05 1.000000e-01 8.500000e-02 PASS | | | | 1 384 21 10 1.980 1.000000e+00 3.181230e-05 1.000000e-01 7.625000e-02 PASS | | | | 1 384 22 10 1.810 1.000000e+00 3.299600e-05 1.000000e-01 8.000000e-02 PASS | | | | 1 384 23 10 1.670 1.000000e+00 2.618960e-05 1.000000e-01 9.125000e-02 PASS | | | | 1 384 24 10 1.530 1.000000e+00 3.136840e-05 1.000000e-01 7.875000e-02 PASS | | | | 1 384 25 10 1.400 1.000000e+00 2.663350e-05 1.000000e-01 8.375000e-02 PASS | | | | 1 384 26 10 1.270 1.000000e+00 3.255210e-05 1.000000e-01 8.625000e-02 PASS | | | | 1 384 27 10 1.140 1.000000e+00 2.692950e-05 1.000000e-01 7.500000e-02 PASS | | | | 1 384 28 10 0.999 1.000000e+00 5.149150e-05 1.000000e-01 9.250000e-02 PASS | | | | 1 384 29 10 0.889 1.000000e+00 3.225620e-05 1.000000e-01 8.750000e-02 PASS | | | | 1 384 30 10 0.772 1.000000e+00 3.536340e-05 1.000000e-01 9.375000e-02 PASS | | | | 1 384 31 10 0.650 1.000000e+00 4.113400e-05 1.000000e-01 9.125000e-02 PASS | | | | 1 384 32 10 0.547 1.000000e+00 3.965440e-05 1.000000e-01 8.750000e-02 PASS | | | | 1 384 33 10 0.428 1.000000e+00 5.489460e-05 1.000000e-01 9.625000e-02 PASS | | | | 1 384 34 10 0.333 1.000000e+00 5.030780e-05 1.000000e-01 8.875000e-02 PASS | | | | 1 384 35 10 0.220 1.000000e+00 3.906250e-05 1.000000e-01 8.875000e-02 PASS | | | | 1 384 36 10 0.127 1.000000e+00 2.929690e-05 1.000000e-01 8.250000e-02 PASS | | | | 1 384 37 10 0.034 1.000000e+00 3.225620e-05 1.000000e-01 9.000000e-02 PASS | | | | 1 384 38 10 -0.066 1.000000e+00 2.737330e-05 1.000000e-01 8.375000e-02 PASS | | | | 1 384 39 10 -0.170 1.000000e+00 2.722540e-05 1.000000e-01 8.500000e-02 PASS | | | | 1 384 40 10 -0.253 1.000000e+00 3.521540e-05 1.000000e-01 7.500000e-02 PASS | | | | 1 384 41 10 -0.344 1.000000e+00 5.563450e-05 1.000000e-01 9.375000e-02 PASS | | | | 1 384 42 10 -0.424 1.000000e+00 2.559780e-05 1.000000e-01 8.750000e-02 PASS | | | | 1 384 43 10 -0.515 1.000000e+00 4.690460e-05 1.000000e-01 9.500000e-02 PASS | | | | 1 384 44 10 -0.605 1.000000e+00 5.755800e-05 1.000000e-01 9.125000e-02 PASS | | | | 1 384 45 10 -0.693 1.000000e+00 3.980230e-05 1.000000e-01 8.000000e-02 PASS | | | | 1 384 46 10 -0.766 1.000000e+00 5.208330e-05 1.000000e-01 9.875000e-02 PASS | | | | 43 TESTS PASSED, 0 TESTS FAILED | +-----------------------------------------------------------------------------------------------------------+

Plots of current SNR values used for BLER testing are shown below:

snr_values.png

The Shared Memory (SHM) configuration file is located at ./cuPHY-CP/gt\_common\_libs/nvlog/config/nvlog\_config.yaml. It contains the following:

Copy
Copied!
            

# nvlog config # name: can see log file at /dev/shm/${name}.log and /tmp/${name}.log # primary: In all processes logging to the same file, set the first starting porcess to be primary, set others to be secondary. # Log levels: -1 - LOG\_NONE, 0 - LOG\_ERROR, 1 - LOG\_CONSOLE, 2 - LOG\_WARN, 3 - LOG\_INFO, 4 - LOG\_DEBUG, 5 - LOG\_VERBOSE # prefix\_opts: refer to nvlog.h # b7 - CPU core number that the caller thread is running on # b6 - Caller thread name # b5 - Caller thread ID # b4 - Global 64-bit sequence number # b3 - Log level: 'E', 'C', 'W', 'I', 'D', 'V' (Error, Console, Warning, Info, Debug, Verbose) # b2 - Module type: 'P', 'S', 'U' (Primary, Secondary, Unknown) # b1 - Date and Time: 1970-01-01 00:00:00.000000 # b0 - Time Only: 00:00:00.000000 nvlog: name: phy shm\_log\_level: 3 # SHM log level console\_log\_level: 1 # Console log level max\_file\_size\_bits: 28 # Size = 2 ^ bits: The rotating log file /var/log/aerial/${name}.log size shm\_cache\_size\_bits: 21 # Size = 2 ^ bits: SHM cache file /dev/shm/${name}.log size log\_buf\_size: 1024 # Max length of one time call of the nvlog API max\_threads: 64 # The MAX thread count which using nvlog togehter save\_to\_file: 1 # Whether to save to a file cpu\_core\_id: -1 # CPU core ID for the background log saving thread.-1 means not bind. prefix\_opts: 0x89 # Log prefix of each line max\_tag\_len: 32 # Max 31 printable characters max\_tag\_num: 1024 # Max tag number nvlog\_tags: - 0: "" # Reserve number 0 for no tag print shm\_level: 3 # Example: overlay shm\_log\_level for a tag console\_level: 1 # Example: overlay console\_log\_level for a tag - 10: "NVLOG" # nvlog - 11: "NVLOG.TEST" - 12: "NVLOG.ITAG"

Note

Max_threads specifies the maximum number of process threads. The default value is 64 threads. To improve performance, the lock-free mechanism doesn’t use Compare-And-Swap, but needs to configure a maximum thread number.

Usage

String Tags

To print with a string tag, define a string TAG in code and use it directly:

Copy
Copied!
            

#define STAG "NVLOG.STAG" NVLOGI(STAG, "This is STAG C printf style log. level=%d\\n",LOG\_INFO); NVSLOGI(STAG) << "This is STAG C++ stream style log. level=" <<LOG\_INFO << "\\n";

To print with a pre-defined integer tag, define an integer TAG in nvlog\_config.yaml:

Copy
Copied!
            

nvlog\_tags: - 10: "NVLOG" # nvlog - 11: "NVLOG.TEST" - 12: "NVLOG.ITAG"

Then use the integer TAG in code:

Copy
Copied!
            

#define NVLOG\_TAG\_BASE\_NVLOG 10 // nvlog #define ITAG (NVLOG\_TAG\_BASE\_NVLOG + 2) // "NVLOG.ITAG" NVLOGI(ITAG, "This is ITAG C printf style log. level=%d\\n", LOG\_INFO); NVSLOGI(ITAG) << "This is ITAG C++ stream style log. level=" << LOG\_INFO << "\\n";

Previous Performance
Next SCF FAPI Support
© Copyright 2022, NVIDIA.. Last updated on Apr 20, 2024.