linux-kernel - Re: [GIT PULL] RISC-V Patches for the 6.5 Merge Window, Part 1

lists.openwall.net		lists / announce owl-users owl-dev john-users john-dev passwdqc-users yescrypt popa3d-users / oss-security kernel-hardening musl sabotage tlsify passwords / crypt-dev xvendor / Bugtraq Full-Disclosure linux-kernel linux-netdev linux-ext4 linux-hardening linux-cve-announce PHC
Open Source and information security mailing list archives

Hash Suite: Windows password security audit tool. GUI, reports in PDF.

[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]

Date:   Fri, 30 Jun 2023 11:47:20 -0700 (PDT)
From:   Palmer Dabbelt <palmer@...osinc.com>
To:     Linus Torvalds <torvalds@...ux-foundation.org>
CC:     linux-riscv@...ts.infradead.org, linux-kernel@...r.kernel.org
Subject:     Re: [GIT PULL] RISC-V Patches for the 6.5 Merge Window, Part 1

On Fri, 30 Jun 2023 09:46:39 PDT (-0700), Linus Torvalds wrote:
> On Fri, 30 Jun 2023 at 08:40, Palmer Dabbelt <palmer@...osinc.com> wrote:
>>
>> There's one merge conflict in MAINTAINERS
>
> Yes, but your suggested resolution is wrong. That "ACPI SERIAL ..."
> entry was moved downwards to be in the right place alphabetically, so
> the entry next to your new RISC-V entry actually needs to be removed.
>
> You seem to just have kept it, so your resolution caused a dup.

Ah, sorry -- I didn't look all that closely because it was just a 
MAINTAINERS update and I knew you just take these all as hints...

>
> The hint from git is the '++' at the beginning of the line:
>
>> ++ACPI SERIAL MULTI INSTANTIATE DRIVER
>> ++M:    Hans de Goede <hdegoede@...hat.com>
>> ++L:    platform-driver-x86@...r.kernel.org
>> ++S:    Maintained
>> ++F:    drivers/platform/x86/serial-multi-instantiate.c
>> ++
>
> which means that git thinks that your merge added those lines from
> thin air (ie they didn't exist in either parent).
>
> NOTE! A '++' line is not a sign of a mis-merge in general. It is
> *purely* a hint. It happens for perfectly fine merges when you either
> end up moving lines around enough that git doesn't see where they came
> from.
>
> Or, perhaps more commonly, when the merge conflict was due to both
> sides changing the same exact same code, and the merge resolution
> comes from neither one directly, but is new code that has combinations
> of both changes.
>
> Anyway, don't worry about it, I appreciate the heads-up regardless and
> I always do my own merge resolutions. I just decided to try to use

... and ya, I figured it was better to just send it anyway.

> this as a "this is how git works" moment.

Awesome, thanks.  I poke around the git merge resolution stuff, but I'm 
never 100% sure so I usually just look at the resulting output files and 
then just paste in whatever git says as a heads up ;)

> Basically, trivial merges where things were just added or removed next
> to each other will normally not result in "++" or "--" lines, so it
> can be a hint that something went wrong if you *thought* your merge
> was trivial, but git gives that kind of output for the end result.
>
> But don't think that "++" means "wrong". It really is just a sign that
> maybe the merge resolution needed a lot of care. And sometimes it's
> entirely a false positive - if git *really* understood merges 100%
> correctly all the time, we'd never need any manual conflict resolution
> at all ;)

OK, I think that makes sense.  Hopefully I can remember next time one 
comes up.

>
>                   Linus