Thread: pgsql: windows: Define WIN32_LEAN_AND_MEAN to make compilation faster.
windows: Define WIN32_LEAN_AND_MEAN to make compilation faster. windows.h includes a lot of other headers, slowing down compilation significantly. WIN32_LEAN_AND_MEAN reduces that a bit. It'd be better to remove the include of windows.h (as well as indirect inclusions of it) from such a central place, but until then... Discussion: https://postgr.es/m/20210921193035.pqzay43vpyv7in43@alap3.anarazel.de Branch ------ master Details ------- https://git.postgresql.org/pg/commitdiff/8162464a25e5314e753c580389f76a9b7f69445b Modified Files -------------- src/include/port/win32_port.h | 8 ++++++++ 1 file changed, 8 insertions(+)
Andres Freund <andres@anarazel.de> writes:
> windows: Define WIN32_LEAN_AND_MEAN to make compilation faster.
I happened to notice from grepping that
src/backend/port/win32/crashdump.c has its own
#define WIN32_LEAN_AND_MEAN
Can't we remove that now? (I kind of wonder what good it did at all,
considering postgres.h has already pulled in <windows.h>.)
regards, tom lane
Hi, On 2021-11-06 18:24:21 -0400, Tom Lane wrote: > Andres Freund <andres@anarazel.de> writes: > > windows: Define WIN32_LEAN_AND_MEAN to make compilation faster. > > I happened to notice from grepping that > src/backend/port/win32/crashdump.c has its own > > #define WIN32_LEAN_AND_MEAN > > Can't we remove that now? (I kind of wonder what good it did at all, > considering postgres.h has already pulled in <windows.h>.) Yea, it seems pointless. There's a semi-related weirdness: Several src/port/ files include <windows.h> themselves (I guess that could be seen as prep for dropping windows.h from win32_port.h). Unless you're already done with that by the time I git pull, I'm gonna remove that redundant WIN32_LEAN_AND_MEAN. Greetings, Andres Freund
Andres Freund <andres@anarazel.de> writes:
> There's a semi-related weirdness: Several src/port/ files include <windows.h>
> themselves (I guess that could be seen as prep for dropping windows.h from
> win32_port.h).
Hmm.
> Unless you're already done with that by the time I git pull, I'm gonna remove
> that redundant WIN32_LEAN_AND_MEAN.
No, I've not done anything --- feel free.
regards, tom lane
On 2021-11-06 15:41:45 -0700, Andres Freund wrote: > Unless you're already done with that by the time I git pull, I'm gonna remove > that redundant WIN32_LEAN_AND_MEAN. Done.