On Tue, Mar 27, 2018 at 12:27 AM, Andreas Müller <[email protected]> wrote: > On Tue, Mar 27, 2018 at 12:11 AM, Martin Jansa <[email protected]> wrote: >> Mostly because I find the proper patch headers useful. >> >> When doing the refresh now, I find it useful to also add original author and >> patch headers even to patches which currently apply cleanly, so that next >> time I can just use "git am foo/*" (or use git as a PATCHTOOL) and get >> usable results. >> >> I understand that it adds unnecessary changes (not needed to fix the >> WARNING), but IMHO better in separate "refresh patches" commit, than next >> time someone uses devtool for actual mozjs upgrade which might refresh the >> patches even more to make them applicable on new base version. >> >> Regards, >> > Understood - but by moving the patches devtool has created to the > original location (often they are already there) and use git add -p > you can decide for each Hunk if that is necessary or not - but it's > not that important to start a longer thread for this :) > Just said 'don't want a longer thread' - and now this :)
An example from V1 demonstrates what I mean (using 'git add -p' forget most of the rest) : -From da3929a96d9c74e11bf37d128890e18fcb745365 Mon Sep 17 00:00:00 2001 +From 08b62f039e0fee50e9399ce2e48b56770f8f71e3 Mon Sep 17 00:00:00 2001 ^ This is not necessary From: Lei Maohui <[email protected]> Date: Mon, 26 Jan 2015 08:53:19 +0900 Subject: [PATCH] mozjs17.0.0: fix the compile bug of powerpc @@ -10,14 +10,16 @@ argument '5' to 'JSBool TryArgumentFormatter(JSContext*, const char**, JSBool, jsval**, __va_list_tag (*)[1])' Signed-off-by: Lei Maohui <[email protected]> + + ^ same here --- - jscpucfg.h | 6 ++++++ + js/src/jscpucfg.h | 6 ++++++ 1 file changed, 6 insertions(+) ^ This is wrong - devtool does not take care for 'S = "${WORKDIR}/${BPN}${PV}/js/src"' in recipe (I should open a bug for that) You'd noticed that by checking in each hunk interactively that's all I wanted to suggest. Devtool is a very strong helper - I use it - but I just don't agree with all it suggests and try to the patch diffs small. Time for zzz.. Andreas -- _______________________________________________ Openembedded-devel mailing list [email protected] http://lists.openembedded.org/mailman/listinfo/openembedded-devel
