Re: Regarding performance issues with the Emacs 25.1 Windows-build

[Noam Postavsky]

On Wed, Nov 9, 2016 at 11:22 AM, Eli Zaretskii <address@hidden> wrote:

>> How was the Windows-version built? Was it built by the same people,
>> using the same setup and the same toolchain? Or was something done
>> differently this time?
>
> The answer is probably that it was not built the same, but I very much
> doubt that the differences could explain a twofold performance hit.
>
> FWIW, my binaries were built by myself using the same compilation
> options (-O2 optimization), but I upgraded my compiler from GCC 4.8.1
> to 5.3.0 between the two versions of Emacs.  Again, I very much doubt
> that this could explain any significant performance differences.

I think the problem is that the 25.1 build is not optimized: I did M-x
report-emacs-bug in each, notice that 25.1 has a setting for CFLAGS
which doesn't include -O2, while 24.5 doesn't have one (and hence uses
the default, which I believe does include -O2).

In GNU Emacs 25.1.1 (x86_64-w64-mingw32)
 of 2016-09-17 built on LAPHROAIG
Configured using:
 'configure --without-dbus --without-compress-install CFLAGS=-static'

In GNU Emacs 24.5.1 (i686-pc-mingw32)
 of 2015-04-11 on LEG570
Windowing system distributor `Microsoft Corp.', version 6.3.9600
Configured using:
 `configure --prefix=/c/usr --host=i686-pc-mingw32'