Chia-Network / bladebit

A high-performance k32-only, Chia (XCH) plotter supporting in-RAM and disk-based plotting
Apache License 2.0
340 stars 107 forks source link

3.1.0 - low-proof Plots (WSL) generated using --disk16 #440

Open jiongxuan opened 7 months ago

jiongxuan commented 7 months ago

Hi, in the last issue(#439 ), I was using UBuntu 22.04 (real system). Then I changed a computer and used Windows 11 (WSL2 Ubuntu 22.04) and GPU drawing to generate compressed Plot, disk 16 mode, and the compression level was C1 .

Compared with directly using UBuntu 22.04, using WSL2 to Plots, the "invalid proof ratio" has decreased, but is still around 24%.

The reason for switching to WSL2 is to eliminate errors without modifying the parameters (BBCU_BUCKET_COUNT) mentioned in #276 . But so far, the effect is not ideal.

Please help me find out what is the reason and how to solve it. Thanks!

My hardware:

GPU: NVidia 2060 Desktop (6Gib vram) RAM: 22 Gib SSD: NVMe 1T System: Windows 11 (WSL2 - Ubuntu 22.04) Cuda: 12.1

Commands and logs: (Keys hidden)

./bladebit_cuda -z 1 -f b81874bf943f54db16cc94687c5081b206100c4c35c98d6d9fef93882d10afda43219b93504008cbdacd7c0fa8336dc3 -c xch109tt48vnhpq2epqufgh5j23tmf047ef4976kl9z5fpee6pq7636q9st380 cudaplot --disk-16 -t1 /mnt/c/Users/aaa/Documents/ChiaTemp /mnt/g/ChiaPlotsNew

Bladebit Chia Plotter Version : 3.1.0 Git Commit : e9836f8bd963321457bc86eb5d61344bfb76dcf0 Compiled With: gcc 9.4.0

[Global Plotting Config] Will create 1 plots. Thread count : 8 Warm start enabled : false NUMA disabled : false CPU affinity disabled : false Farmer public key : b818... Pool contract address : xch10... Compression Level : 1 Benchmark mode : disabled Warning: 16G mode is experimental and still under development. Please use the --check parameter to validate plots when using this mode. Direct I/O not supported in 16G mode at the moment. Disabing it.

[Bladebit CUDA Plotter] Host RAM : 11 GiB Plot checks : disabled

Selected cuda device 0 : NVIDIA GeForce RTX 2060 CUDA Compute Capability : 7.5 SM count : 30 Max blocks per SM : 16 Max threads per SM : 1024 Async Engine Count : 2 L2 cache size : 3.00 MB L2 persist cache max size : 0.00 MB Stack Size : 1.00 KB Memory: Total : 6.00 GB Free : 5.00 GB

Allocating buffers (this may take a few seconds)... Kernel RAM required : 4828776144 bytes ( 4605.08 MiB or 4.50 GiB ) Intermediate RAM required : 4378927104 bytes ( 4176.07 MiB or 4.08 GiB ) Host RAM required : 2147483648 bytes ( 2048.00 MiB or 2.00 GiB ) Total Host RAM required : 6976259792 bytes ( 6653.08 MiB or 6.50 GiB ) GPU RAM required : 6161465344 bytes ( 5876.03 MiB or 5.74 GiB ) Allocating buffers... Done.

Generating plot 1 / 1: 8a4ee1502ac58ce5f4035892fcf4014a447f773e667574a17041f9705967ebde Plot temporary file: /mnt/g/ChiaPlotsNew/plot-k32-c01-2023-11-16-13-45-8a4ee1502ac58ce5f4035892fcf4014a447f773e667574a17041f9705967ebde.plot.tmp

Generating F1 Finished F1 in 176.83 seconds. Table 2 completed in 476.27 seconds with 4294769045 entries. Table 3 completed in 615.28 seconds with 4294632684 entries. Table 4 completed in 727.03 seconds with 4294236853 entries. Table 5 completed in 735.84 seconds with 4293510990 entries. Table 6 completed in 679.44 seconds with 4292085654 entries. Table 7 completed in 575.26 seconds with 4289181417 entries. Finalizing Table 7 Finalized Table 7 in 245.20 seconds. Completed Phase 1 in 4231.16 seconds Marked Table 6 in 50.68 seconds. Marked Table 5 in 46.70 seconds. Marked Table 4 in 49.25 seconds. Marked Table 3 in 50.33 seconds. Completed Phase 2 in 196.95 seconds Compressing Table 2 and 3... Step 1 completed step in 468.68 seconds. Step 2 completed step in 322.67 seconds. Completed table 2 in 791.34 seconds with 3439308024 / 4294632684 entries ( 80.08% ). Compressing tables 3 and 4... Step 1 completed step in 401.96 seconds. Step 2 completed step in 448.71 seconds. Step 3 completed step in 257.21 seconds. Completed table 3 in 1107.88 seconds with 3465069278 / 4294236853 entries ( 80.69% ). Compressing tables 4 and 5... Step 1 completed step in 245.69 seconds. Step 2 completed step in 437.72 seconds. Step 3 completed step in 253.68 seconds. Completed table 4 in 937.09 seconds with 3531097905 / 4293510990 entries ( 82.24% ). Compressing tables 5 and 6... Step 1 completed step in 232.34 seconds. Step 2 completed step in 419.37 seconds. Step 3 completed step in 235.41 seconds. Completed table 5 in 887.11 seconds with 3710423390 / 4292085654 entries ( 86.45% ). Compressing tables 6 and 7... Step 1 completed step in 232.02 seconds. Step 2 completed step in 413.19 seconds. Step 3 completed step in 243.72 seconds. Completed table 6 in 888.94 seconds with 4289181417 / 4289181417 entries ( 100.00% ). Serializing P7 entries Completed serializing P7 entries in 116.58 seconds. Completed Phase 3 in 4728.95 seconds Completed Plot 1 in 9157.06 seconds ( 152.62 minutes )

/mnt/g/ChiaPlotsNew/plot-k32-c01-2023-11-16-13-45-8a4ee1502ac58ce5f4035892fcf4014a447f773e667574a17041f9705967ebde.plot.tmp -> /mnt/g/ChiaPlotsNew/plot-k32-c01-2023-11-16-13-45-8a4ee1502ac58ce5f4035892fcf4014a447f773e667574a17041f9705967ebde.plot Completed writing plot in 0.24 seconds Final plot table pointers: Table 1: 0 ( 0x0000000000000000 ) Table 2: 1288476952 ( 0x000000004ccc9918 ) Table 3: 15287538552 ( 0x000000038f355378 ) Table 4: 29372847477 ( 0x00000006d6c21575 ) Table 5: 43726554402 ( 0x0000000a2e4e4522 ) Table 6: 58809214977 ( 0x0000000db14d6801 ) Table 7: 76244487252 ( 0x00000011c0868c54 ) C 1 : 4096 ( 0x0000000000001000 ) C 2 : 1719776 ( 0x00000000001a3de0 ) C 3 : 1719952 ( 0x00000000001a3e90 )

Final plot table sizes: Table 1: 0.00 MiB Table 2: 13350.55 MiB Table 3: 13432.80 MiB Table 4: 13688.76 MiB Table 5: 14383.95 MiB Table 6: 16627.57 MiB Table 7: 16873.24 MiB C 1 : 1.64 MiB C 2 : 0.00 MiB C 3 : 1227.15 MiB

./bladebit_cuda check -n 1000 /mnt/g/ChiaPlotsNew/plot-k32-c01-2023-11-16-13-45-8a4ee1502ac58ce5f4035892fcf4014a447f773e667574a17041f9705967ebde.plot Checking 1000 random proofs with seed 0xcbbc70b37019beba1d3856b4f6cd4965ece22c5554e75fcf7af54f7bacc90f51... Plot compression level: 1 Using GPU for decompression. 10%... 20%... 30%... 40%... 50%... 60%... 70%... 80%... 90%... Seed used: 0xcbbc70b37019beba1d3856b4f6cd4965ece22c5554e75fcf7af54f7bacc90f51 Proofs requested/fetched: 703 / 1000 ( 70.300% ) Proof fetches failed : 10 ( 1.000% ) Proof validation failed : 243 ( 24.300% )

teamwest93 commented 7 months ago

well, proof rate - pure random. With wsl I saw 4/30 PR and 32/30. But always have some numbers of PFF and PVF