Re: [PULL v7 000/151] Meson-based build system

[Paolo Bonzini]

On 19/08/20 23:59, Eric Blake wrote:
> On 8/19/20 4:32 PM, Paolo Bonzini wrote:
>> The following changes since commit
>> d0ed6a69d399ae193959225cdeaa9382746c91cc:
>>
>>    Update version for v5.1.0 release (2020-08-11 17:07:03 +0100)
>>
>> are available in the Git repository at:
>>
>>    https://gitlab.com/bonzini/qemu.git tags/for-upstream
> 
> Unrelated to the pull request proper, but I note in MAINTAINERS that you
> have listed:
> 
> T: git https://github.com/bonzini/qemu.git scsi-next
> 
> Have you switched to gitlab as your personal mirroring preference, and
> if so, should MAINTAINERS be updated?

Yes, but also I've not used a scsi-next branch for years. :)

>> for you to fetch changes up to 2eddb3c65821dce76433d5da6f3e6419349d1b77:
>>
>>    docs: convert build system documentation to rST (2020-08-19
>> 16:13:30 -0400)
>>
>> v6->v7:
>> * new patch to preserve compatibility symlinks from previous binary
>> locations
>> * fixed cut-and-paste error in linux-user/mips/meson.build
>> * preserve compatibility check-block target even if no block tests are
>> defined
> 
> FWIW: I haven't done any review or testing of the earlier iterations of
> this series.  But since it will be landing soon, I merged this tag to an
> incremental build tree that was previously sitting on a finished 5.1
> build (sources in qemu/, VPATH build in qemu/build/), typed 'make' then
> 'make check', and everything finished successfully at least for my
> typical setup.  So good job on getting this massive rewrite in while
> still preserving at least the common 'make' interface.

I must admit I was a bit lazy with incremental builds until Peter
pressed me for it, and I must say it wasn't _that_ hard to do, with a
fresh mind.  I tried a bunch of touch&&make scenarios and they all
worked fine.  It always helps to steal tricks from Autoconf's bag, since
QEMU's "./config.status --skip-meson" is basically the same as
Autoconf's "./config.status --recheck".

Paolo