Re: [PATCH v22 06/17] meson: add target_user_arch

[Eric Blake]

On 2/24/21 3:21 PM, Philippe Mathieu-Daudé wrote:
> On 2/24/21 2:34 PM, Claudio Fontana wrote:
>> the lack of target_user_arch makes it hard to fully leverage the
>> build system in order to separate user code from sysemu code.
>>
>> Provide it, so that we can avoid the proliferation of #ifdef
>> in target code.
>>
>> Signed-off-by: Claudio Fontana <cfontana@suse.de>
>> Reviewed-by: Alex Bennée <alex.bennee@linaro.org>
>>
>> [claudio: added changes for new target hexagon]
> 
> Again, this line goes ...
> 
>> ---
> 
> ... here. Else it is not stripped by git-am and ends
> burried in the repository (see commit 940e43aa30e).

If you are modifying a patch originally written by someone else (that
is, their S-o-b appears first, but your edits mean you also add S-o-b),
it is courteous to include your modifications in the commit log in this
manner.  (For an example, see commit 2c4c556e06)

You're right that it can look fishy if your changlog appears on your own
commit (if you rebased things with no one else touching the patch in
between, just update the commit message as part of that rebase; the
changelog goes after the --- for review in that case).  But it's not
completely wrong: you'll see me doing it when wearing my maintainer hat
and preparing a pull request, and modifying my own patch different from
how it was posted on the mailing list while wearing my developr hat
prior to the pull request (see commit c930831446 for an example)

-- 
Eric Blake, Principal Software Engineer
Red Hat, Inc.           +1-919-301-3226
Virtualization:  qemu.org | libvirt.org