[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]
[bug #55526] FTBFS on GNU/kFreeBSD: NSWorkspace.m:1252:27: error: 'MOUNT
From: |
Yavor Doganov |
Subject: |
[bug #55526] FTBFS on GNU/kFreeBSD: NSWorkspace.m:1252:27: error: 'MOUNTED_PATH' undeclared |
Date: |
Mon, 21 Jan 2019 18:38:21 -0500 (EST) |
User-agent: |
Mozilla/5.0 (X11; Linux x86_64) AppleWebKit/537.36 (KHTML, like Gecko) Chrome/72.0.3626.53 Safari/537.36 |
Follow-up Comment #4, bug #55526 (project gnustep):
Thanks; I won't be able to build-test it until the next gnustep-gui upload to
Debian.
GNU/kFreeBSD is more closer to GNU/Linux than to FreeBSD, because the
underlying C library (glibc) is the same. This is basically a GNU system but
running with another kernel. Part of the confusion comes because people talk
about Linux as an OS when in fact it's just a kernel.
For programming purposes, you can (and should) treat GNU/kFreeBSD as a glibc
platform -- you are already using the portable macros as documented in the
glibc manual (albeit confusingly redefined to the deprecated ones) so that
should be sufficient.
The conditionals in -mountedLocalVolumePaths are of no concern since the
HAVE_GETMNTINFO branch will be compiled on GNU/kFreeBSD, which is fine. It is
also fine to compile the other branch (HAVE_GETMNTENT) and you can enforce it
on GNU/kFreeBSD if you wish so with the appropriate defines.
But the result should be the same. When Bruno Haibble ported the GNU C
Library to the FreeBSD kernel (hats off to this hacker!), he implemented
interfaces which were not feasible with the Linux kernel but were expected and
customary with a FreeBSD kernel, like getmntinfo.
So there it is, a hybrid system which proves the defines were wrong as it is
perfectly possible to have both.
_______________________________________________________
Reply to this item at:
<https://savannah.gnu.org/bugs/?55526>
_______________________________________________
Message sent via Savannah
https://savannah.gnu.org/