Rich-T-kid commented on PR #10262: URL: https://github.com/apache/arrow-rs/pull/10262#issuecomment-4895927359
[main] <img width="2428" height="776" alt="Image 7-6-26 at 1 20 PM" src="https://github.com/user-attachments/assets/8d2de995-0714-47e9-afb6-65ce179261ad"; /> [this branch] <img width="2430" height="448" alt="Image 7-6-26 at 1 20 PM (1)" src="https://github.com/user-attachments/assets/1b38263a-f873-4789-bbab-d113775788e5"; /> Notably the percentage of calls to malloc/ time taken by malloc are cut in half, which is expected since we are doing fewer, large calls upfront. 🤔 There does seem to be a large difference between `alloc::raw_vec::RawVecInner::finish_grow` [main] <img width="2374" height="108" alt="Image 7-6-26 at 1 34 PM" src="https://github.com/user-attachments/assets/14e032e3-5be6-4ad6-8b04-0375705c0a5f"; /> [this branch] <img width="2440" height="170" alt="Image 7-6-26 at 1 35 PM" src="https://github.com/user-attachments/assets/36a20d56-e1c1-4998-bd4b-773b9ef3d0a3"; /> [main] <img width="3008" height="1346" alt="Image 7-6-26 at 1 46 PM" src="https://github.com/user-attachments/assets/b1ed6c4d-7d0e-469e-a127-4cee6310c1bf"; /> [my branch] <img width="3010" height="1178" alt="Image 7-6-26 at 1 47 PM" src="https://github.com/user-attachments/assets/a0d426b8-bf63-4477-bc1a-a97dc22f2672"; /> Show a decrease in malloc allocations that get filled in with `memmove` -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected]
