Quoting Lynne (2023-05-11 19:12:56) > May 11, 2023, 18:34 by an...@khirnov.net: > > > Quoting Lynne (2023-04-24 17:56:38) > > > >> From 6b5301aa29b63b90d04505c9386822b2e207a038 Mon Sep 17 00:00:00 2001 > >> From: Lynne <d...@lynne.ee> > >> Date: Thu, 29 Dec 2022 21:16:21 +0100 > >> Subject: [PATCH 55/97] vulkan: rewrite to support all necessary features > >> > >> --- > >> libavutil/vulkan.c | 2145 ++++++++++++++++++---------------- > >> libavutil/vulkan.h | 515 ++++---- > >> libavutil/vulkan_functions.h | 1 + > >> 3 files changed, 1344 insertions(+), 1317 deletions(-) > >> > > > > lol > > > > We stopped doing development like this 15 years ago. > > > > First, I'm criticized for having too many patches. Now, I'm criticized for > having too few. There's no winning.
I have no issue with too many patches - my work on ffmpeg CLI is over 200 commits this year alone. Small patches are reviewable, this horrorshow is not. > This touches EVERYTHING, and so it's titled appropriately. If you dislike > the word in the commit message, I can change it to something with > more marketing impact. No, I dislike the fact that there's a giant uberpatch that rewrites everything with zero explanation or justification. Ideally, it should be a series of small individually reviewable changes. If that is for some reason not feasible, the commit message should contain * good justification for why it is not feasible * detailed explanation on what exactly is being done and why -- Anton Khirnov _______________________________________________ ffmpeg-devel mailing list ffmpeg-devel@ffmpeg.org https://ffmpeg.org/mailman/listinfo/ffmpeg-devel To unsubscribe, visit link above, or email ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
