bpo-46740: speedup telnetlib's buffer transfers #31449
Closed
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
This patch leverages the very low frequency of IAC (control) characters: instead of reading/interpreting/appending per character, it's much faster to search them linearly (if any), then copy the data (non-IAC) slice to the cooked queue.
rawq
index handling in_next_nonIAC_slice
is almost copied fromrawq_getchar
.Also calls
socket.recv
with the currently recommended buffer size, 4096.Patching
process_rawq
patching was inspired by point 2) in this very old Python-dev topichttps://python-dev.python.narkive.com/A86fs3PG/patch-to-telnetlib-py
and yields a 5x speedup too.
https://bugs.python.org/issue46740