mirabilos

Unfortunately, http://www.mirbsd.org/wtf.cgi?q=RAQM Sure. [x.zip](https://github.com/libgd/libgd/files/7965225/x.zip) ```php

Pierre Joye dixit: ***@***.*** RAQM is an optional library for complex text layout >@(vertical, RTL and co), https://github.com/HOST-Oman/libraqm Ah, okay. I thought it was some acronym I didn’t know yet....

Pierre Joye dixit: >but 2.3.x yes, which may cause the regression. Is that a reply to… >> 2.2.5-5.2 doesn’t even show anything to do with raqm. … this? Because 2.3.0-2...

Pierre Joye dixit: >I mean it could have been due to raqm. But it is not used. I will >investigate that next week. Thanks!

Uhm, careful ping?

Judging from [the latest kfreebsd-amd64 build log](https://buildd.debian.org/status/fetch.php?pkg=performous&arch=kfreebsd-amd64&ver=1.1%2Bgit20190701.9928c27-2&stamp=1564507585&raw=0), there’s also no OS definition for GNU/kFreeBSD (the FreeBSD kernel with GNU userland). The build fails because neither `BOOST_OS_BSD` nor `BOOST_OS_LINUX` are defined…...

`x32` is basically a separate architecture, yes. It shares the x86 common architecture, and the `amd64` (what the GNU people call `x86_64`) CPU architecture, but uses a different ABI… kinda...

(Ideally you would not pass *any* of `-m32`, `-m64`, `-mx32` but let the compiler use the natively selected one.)

jerch dixit: ***@***.*** Why would it be needed to treat this separately, why not just use existing API: > >```typescript >pty.spawn('/bin/sh', ['-sh', ...], ...) >``` AIUI in the existing API,...

Hi jerch, ***@***.*** Thx, found the piece of code, where it gets copied over as ***@***.***[0]. Is that '-' convention of `login` specced somewhere (maybe in >@POSIX?) or at least...