Created attachment 528202 [details, diff] dont-run-previous-iteration-commands-again-9c32bdab29b3.patch ki18n-5.45.0 shipped with a crippling performance problem. See https://phabricator.kde.org/D12216. The end result looks like this: # emerge -1 =kde-plasma/plasma-workspace-5.12.4-r1 . . . [ 1%] Generating mo... And here we stay for longer than I am willing to wait for it to finish. Meanwhile, a sampling of the number of zombie msgfmt processes on my machine shows highly varying numbers, but averaging around maybe 2000 zombies, occasionally many more than that. The attached =kde-frameworks/ki18n-5.45.0-r1 patch, taken from https://phabricator.kde.org/file/data/vvacwo6tnxe45lrwiqnp/PHID-FILE-7ldqftbs546soi6ao6aq/D12216.diff without modification, once applied and merged, fixes the problem for all its consumers including plasma-workspace (which just happens to be particularly impacted due to the high amount of localized files in it). The patch is upstream so it's technically a backport.
The bug has been closed via the following commit(s): https://gitweb.gentoo.org/repo/gentoo.git/commit/?id=4b2a7ee17866dd768651b51cd6640f39506a8469 commit 4b2a7ee17866dd768651b51cd6640f39506a8469 Author: Andreas Sturmlechner <asturm@gentoo.org> AuthorDate: 2018-04-22 08:00:46 +0000 Commit: Andreas Sturmlechner <asturm@gentoo.org> CommitDate: 2018-04-22 09:32:37 +0000 kde-frameworks/ki18n: Fix performance of build-pofiles Thanks-to: Greg Turner <gmturner007@ameritech.net> KDE-Bug: https://bugs.kde.org/show_bug.cgi?id=393141 See also: https://phabricator.kde.org/D12216 Closes: https://bugs.gentoo.org/653756 Package-Manager: Portage-2.3.31, Repoman-2.3.9 .../files/ki18n-5.45.0-fix-build-pofiles.patch | 38 ++++++++++++++++++++ kde-frameworks/ki18n/ki18n-5.45.0-r2.ebuild | 40 ++++++++++++++++++++++ 2 files changed, 78 insertions(+)
