Make snappy decompress check more efficient

[cheinger]

• Add status checking of decompressed snappy parquet pages when using cuDF snappy decompression.
• Improve the performance of checking decompressed snappy parquet pages when using nvCOMP snappy decompression by reducing the number of synchronizes.

[GPUtester]

Can one of the admins verify this patch?

[devavret]

ok to test

[devavret]

I realized we allocate and pass comp_stat to snappy_decompress() but don't use it. We should use it for checking instead.

[devavret]

@cheinger Also, please don't force-push. It decouples the review comments from the code and makes it hard to see what changed in the most recent commit

[devavret]

@cheinger is this ready for another review? Click the re-request review button next to my name in reviewers when ready.

[vyasr]

This should probably be retargeted to 22.04 at this point.

[devavret]

Nearly there.

[jbrennan333]

@cheinger what is the status of this? Looks like it might need to be merged up to head of branch-22.04.
Would be nice to get this in for branch-22.04 if that is still possible.

[cheinger]

@jbrennan333 Hey Jim - yeah I would like to get this in as well, however I keep running into issues that don't seem to be related to these changes? 🤔 I've rebased several times but always run into a different issue. Maybe you could provide some guidance? Thanks

[jbrennan333]

I know there was a bug earlier this week that was impacting pre-commit tests which was fixed. This happens sometimes. It's a big project with a lot of changes going in - once you commit, you should keep an eye on those pre-commit checks and follow up quickly. I think this one has missed the opportunity to get into 22.04 as we are already in burn-down. Would be great to get it in early in 22.06.

[cheinger]

rerun tests

[devavret]

rerun tests

[cheinger]

rerun tests

[codecov]

Codecov Report

Merging #9995 (f331301) into branch-22.04 (4596244) will decrease coverage by 0.00%.
The diff coverage is 93.70%.

@@               Coverage Diff                @@
##           branch-22.04    #9995      +/-   ##
================================================
- Coverage         86.13%   86.13%   -0.01%     
================================================
  Files               139      139              
  Lines             22438    22435       -3     
================================================
- Hits              19328    19324       -4     
- Misses             3110     3111       +1     

Impacted Files	Coverage Δ
python/cudf/cudf/core/dataframe.py	93.57% <ø> (ø)
python/cudf/cudf/core/series.py	95.16% <ø> (ø)
python/dask_cudf/dask_cudf/tests/test_accessor.py	98.41% <ø> (ø)
python/cudf/cudf/core/column/decimal.py	91.30% <73.68%> (-1.01%)	⬇️
python/cudf/cudf/core/column/categorical.py	89.63% <84.61%> (-0.29%)	⬇️
python/cudf/cudf/core/column/string.py	88.91% <94.44%> (+0.64%)	⬆️
python/cudf/cudf/core/column/numerical.py	95.33% <95.83%> (+0.35%)	⬆️
python/cudf/cudf/api/types.py	89.79% <100.00%> (ø)
python/cudf/cudf/core/column/column.py	89.27% <100.00%> (+0.10%)	⬆️
python/cudf/cudf/core/column/datetime.py	88.55% <100.00%> (-0.52%)	⬇️
... and 21 more

---

Continue to review full report at Codecov.

Legend - Click here to learn more
Δ = absolute <relative> (impact), ø = not affected, ? = missing data
Powered by Codecov. Last update 621d26f...f331301. Read the comment docs.

[PointKernel]

@gpucibot merge