bulk88

PostgreSQL found this exact performance problem on Windows OS (WinSock.dll problem) and on MSVC (the &

> The idea of the change sounds like a very good idea, but I can't help wondering if it can't be wired more simply. What are your thoughts on dropping...

https://github.com/gcc-mirror/gcc/commit/73984f8494f31ba93aaebf84a50dc77fd4017c11 GCC added pattern matching the DIYed `htonl()` macro only recently in 2014 if above commit is true

> I agree with @iabyn about commit messages. And I want to amplify on that. > > A commit message is a permanent record. It will be read many many...

> [@bulk88](https://github.com/bulk88), I recommend you write a succinct version of this bug report and post it [here](https://github.com/Perl-Toolchain-Gang/ExtUtils-MakeMaker/issues). This is a revert needed in core issue at this point. Its not...

> I've just built 32-bit Windows build of current blead (at commit [79d4169](https://github.com/Perl/perl5/commit/79d4169d029dd93108c936e14e4ce01b1bd0973c)), using gcc-15.1.0 (from https://winlibs.com) and can't find any issue. Here is perl -V: IDK how Winlibs.com configs/bundles...

I 've never CC_HOME env var in the root makefile, my GCCs can find their /lib folder automatically. changing Win32.pms makefile by hand, top is from current blead to bottom...

https://github.com/Perl-Toolchain-Gang/ExtUtils-MakeMaker/issues/277 might be related but bug ticket was opened in 2018

> NST_TOP=D:\perl-5.41.13-830 CCTYPE=GCC CCHOME=C:\sp\_32\sp-5.32.1-64int\c USE_MINGW_ANSI_STDIO=define > > Use the i386 GCC that came with i386 Straw 5.32. > > Building blead (at commit [79d4169](https://github.com/Perl/perl5/commit/79d4169d029dd93108c936e14e4ce01b1bd0973c)) using that compiler, I start with:...

Edit: I forgot to include ` SKIP_CCHOME_CHECK=1` in `gmake SKIP_CCHOME_CHECK=1 test`. I set `SKIP_CCHOME_CHECK=1` as a global ENV var on my dev capable systems. 1st post continues below. Yes @sisyphus...