fix: Set memory decompression limit

[mattwittwer]

What does the PR do?

This PR adds additional size validation to compressed inputs.

Checklist

• PR title reflects the change and is of format <commit_type>: <Title>
• Changes are described in the pull request.
• Related issues are referenced.
• Populated github labels field
• Added test plan and verified test passes.
• Verified that the PR passes existing CI.
• Verified copyright is correct on all changed files.
• Added succinct git squash message before merging ref.
• All template sections are filled out.
• Optional: Additional screenshots for behavior/output changes with before/after.

Commit Type:

Check the conventional commit type
box here and add the label to the github PR.

• build
• ci
• docs
• feat
• fix
• perf
• refactor
• revert
• style
• test

Related PRs:

Where should the reviewer start?

The main changes are in: src/data_compressor.h please review the logic to calculate the size of the decompressed buffer and then review the testing changes.

Test plan:

Yes specific tests for this decompression function were added here: src/test/data_compressor_test.cc
And additional end to end tests were added here: qa/L0_http/http_input_size_limit_test.py

• CI Pipeline ID:

40402527

Caveats:

Background

Related Issues: (use one of the action keywords Closes / Fixes / Resolves / Relates to)

• closes GitHub issue: #xxx

[yinggeh]

Please also update command_line_parser.cc.

[mattwittwer]

Please also update command_line_parser.cc.

Updated the description of http-max-input-size in command_line_parser.cc.

[whoisj]

LGTM

[mattwittwer]

Encountered error with base case where decompression size == decompression limit failed to allocate:

[ RUN      ] DataCompressorTest.DecompressionSizeLimitGzip
/opt/tritonserver/server/src/test/data_compressor_test.cc:698: Failure
Value of: (err == nullptr)
  Actual: false
Expected: true
64 MB data should decompress at exact 64 MB limit: Decompressed data size exceeds the maximum allowed value of 67108864 bytes. Use --http-max-input-size to increase the limit.
[  FAILED  ] DataCompressorTest.DecompressionSizeLimitGzip (1476 ms)
[ RUN      ] DataCompressorTest.DecompressionSizeLimitDeflate
/opt/tritonserver/server/src/test/data_compressor_test.cc:776: Failure
Value of: (err == nullptr)
  Actual: false
Expected: true
64 MB data should decompress at exact 64 MB limit: Decompressed data size exceeds the maximum allowed value of 67108864 bytes. Use --http-max-input-size to increase the limit.
[  FAILED  ] DataCompressorTest.DecompressionSizeLimitDeflate (1510 ms)
[----------] 12 tests from DataCompressorTest (5481 ms total)

[----------] Global test environment tear-down
[==========] 12 tests from 1 test suite ran. (5481 ms total)
[  PASSED  ] 10 tests.
[  FAILED  ] 2 tests, listed below:

Resolved with an additional check:
If there is nothing more to decompress. Break the inflate() loop. No need to generate a new buffer. This is especially important when there is no more room to allocate additional buffer.
https://github.com/triton-inference-server/server/blob/mwittwer/set_memory_decompression_limit/src/data_compressor.h#L314-L318

[yinggeh]

LGTM