mirror of
https://git.freebsd.org/ports.git
synced 2025-06-10 07:10:32 -04:00
I completely reworked these unmaintained ports. I turned mess into a proper slave port as the differences from mame were few. Some notes: * converted to GitHub (this saved messing with the double-zip / dos2unix per file arrangement, no more post-extract target) * got rid of patchlevel support accordingly * got rid of DIST_SUBDIR accordingly * Clang from FreeBSD 10 cannot build this! It fails will some kind of signal. I did not test on FreeBSD 11 yet. For now, degrade the ports to use GCC always. Note the makefiles were hardcoded to GCC. I fixed the hardcoding so it uses environment but perhaps it's known that mame/mess is now gcc-only * Some options are no longer optional, e.g. BUILTIN_DEBUGGER, SDL2 These were made unconditional accordingly. * mess is now a minimal slave port. There are some additions to mame Makefile to support the slave. * I used PORTDOCS and PORTEXAMPLE to simplify each pkg-plist * DragonFly support still pending. This port discovered a bug in GCC 5.1 release which has been fixed recently, so the base compiler has to be upgraded before testing can resume.
11 lines
426 B
C
11 lines
426 B
C
--- 3rdparty/lzma/C/CpuArch.h.orig 2015-05-24 15:52:12 UTC
|
|
+++ 3rdparty/lzma/C/CpuArch.h
|
|
@@ -16,7 +16,7 @@ MY_CPU_LE_UNALIGN means that CPU is LITT
|
|
If MY_CPU_LE_UNALIGN is not defined, we don't know about these properties of platform.
|
|
*/
|
|
|
|
-#if defined(_M_X64) || defined(_M_AMD64) || defined(__x86_64__)
|
|
+#if defined(_M_X64) || defined(_M_AMD64) || defined(__x86_64__) || defined(__amd64__)
|
|
#define MY_CPU_AMD64
|
|
#endif
|
|
|