One of the busiest functions in rpm was getting passed a new copy of a big unordered_map on each call during fingerprinting because "somebody" forgot a & in commit 08a6a5e848f007c2e8d83fc1df8712d1a43229f4. This isn't even noticeable in the test-suite or daily "update a few dozen packages" operation but when attempting to install/update a few thousand packages dragged this little buglet to the light...
Why oh why C++ do you behave in such an idiotic manner in the face of stupid mistakes. Raw pointers at least give you a compiler error. You can view, comment on, or merge this pull request online at: https://github.com/rpm-software-management/rpm/pull/3422 -- Commit Summary -- * Fix a horrible performance regression caused by missing & -- File Changes -- M lib/fprint.cc (2) -- Patch Links -- https://github.com/rpm-software-management/rpm/pull/3422.patch https://github.com/rpm-software-management/rpm/pull/3422.diff -- Reply to this email directly or view it on GitHub: https://github.com/rpm-software-management/rpm/pull/3422 You are receiving this because you are subscribed to this thread. Message ID: <rpm-software-management/rpm/pull/3...@github.com>
_______________________________________________ Rpm-maint mailing list Rpm-maint@lists.rpm.org http://lists.rpm.org/mailman/listinfo/rpm-maint
