Commit Graph

119828 Commits

Author SHA1 Message Date
simonb 35d9e54664 If the make variable "USE_BINUTILS" is yet, use correct options to link
a kernel with recent(ish) binutils ld.
2003-10-27 07:39:10 +00:00
simonb c0a89c7e21 Note the GCC PR for the Duff's device warning. 2003-10-27 07:31:07 +00:00
junyoung 592280de7a index() shouldn't be used in the kernel. Use strchr() instead. 2003-10-27 07:28:15 +00:00
simonb 075b65698a Use -Wno-error to compile a file with Duff's device and document this as
a hack.
2003-10-27 07:26:17 +00:00
chs b3a3d69b22 uninitialized variables. 2003-10-27 07:14:25 +00:00
chs 3b6d3c710c uninitialized variables 2003-10-27 07:07:34 +00:00
lukem 7568eb663a skip a leading `-' on the arg0 name when calling el_init() 2003-10-27 06:19:29 +00:00
christos 50459b4529 Fix uninitialized variable warnings. 2003-10-27 06:15:52 +00:00
christos 577a829bf1 fix make -jn by adding explicit dependency of in_cksum.o to assym.h 2003-10-27 05:49:15 +00:00
christos 63b3407c02 add posix semaphores 2003-10-27 05:34:48 +00:00
simonb 499ef1c1bb Remove (the now non-compiling) support for 128bit FP emulation, which
isn't needed for PowerPC anyway.
2003-10-27 04:32:02 +00:00
simonb 354075932d Fix a bogus uninitialised warning. 2003-10-27 04:30:32 +00:00
christos 7f7c7b7d99 Back out the libretto related hacks I accidentally committed. They are
in CVS in case one wants them :-)
2003-10-27 04:10:43 +00:00
simonb 1baeb1502a Keep bogus gcc -Wuninitialised warnings happy. 2003-10-27 03:58:17 +00:00
lukem b81744894f appease gcc's uninitialised variable detection 2003-10-27 03:51:35 +00:00
lukem 5dfdeb16e3 pull in nbtool_config.h #if HAVE_NBTOOL_CONFIG_H 2003-10-27 03:46:15 +00:00
mrg dcce429e93 fix a typo that caused the build to fail. mmm commit before compile 2003-10-27 03:09:43 +00:00
simonb 6836656005 Remove assigned-to but otherwise unused variables. 2003-10-27 02:58:31 +00:00
christos 5f4474073a Explain the stack layout correctly now that that I understand it :-) 2003-10-27 02:33:06 +00:00
simonb ecb9893b78 Use "#if" to compare two tokens, not "#ifdef". 2003-10-27 02:30:26 +00:00
lukem 477697b1f2 Overhaul how `build.sh tools' are used:
*	Rename "config.h" to "nbtool_config.h" and
	HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
	This makes in more obvious in the source when we're using
	tools/compat/config.h versus "standard autoconf" config.h

    *	Consistently move the inclusion of nbtool_config.h to before
    	<sys/cdefs.h> so that the former can provide __RCSID() (et al),
    	and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
	#if HAVE_NBTOOL_CONFIG_H
	#include "nbtool_config.h"
	#endif
to the top of the source files (for the general case).
2003-10-27 02:17:18 +00:00
simonb 8338dcf797 More bogus uninitialised warnings. 2003-10-27 02:16:15 +00:00
cl 7585316531 initialize uninitialized 2003-10-27 02:03:10 +00:00
uwe e921d39cf4 Note sh3el switch to GDB 5.3. 2003-10-27 01:46:04 +00:00
mrg 3b4ce5498c turn on GCC3 for armeb 2003-10-27 01:45:20 +00:00
lukem 444edb22a9 Overhaul how `build.sh tools' are used:
*	Rename "config.h" to "nbtool_config.h" and
	HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
	This makes in more obvious in the source when we're using
	tools/compat/config.h versus "standard autoconf" config.h

    *	Consistently move the inclusion of nbtool_config.h to before
	<sys/cdefs.h> so that the former can provide __RCSID() (et al),
	and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
	#if HAVE_NBTOOL_CONFIG_H
	#include "nbtool_config.h"
	#endif
to the top of the source files (for the general case).
2003-10-27 01:42:50 +00:00
mrg 8d8c032590 add armeb-netbsdelf GCC3 support. (not tested) 2003-10-27 01:37:08 +00:00
uwe 1312264774 Mark GCC3 for sh3el as "mostly working". Remaining issues are
toolchain/22452 and toolchain/22489 (gcc's optimization/11937).
2003-10-27 01:35:12 +00:00
uwe 0f1f83b0cc Switch sh3el to GDB 5.3. 2003-10-27 01:29:58 +00:00
simonb eeb03b3e1b "Fix" bogus gcc3 uninitialised warning. 2003-10-27 01:17:59 +00:00
kleink 964d6747b5 Err, rename some members added in previous to make them reflect their
semantics better.
2003-10-27 01:11:46 +00:00
cl 16d49327a2 appease lint's ``warning: nonportable bit-field type'' in MD header 2003-10-27 01:05:08 +00:00
wiz 51677832e7 Close parentheses. 2003-10-27 00:57:04 +00:00
christos 6664eb5f2d Add __sigtramp2 2003-10-27 00:46:05 +00:00
christos d9676dda89 Signal trampoline stuff for siginfo.
XXX: I don't understand why I am off by 48 bytes.
2003-10-27 00:45:36 +00:00
cl 44f9690b65 g/c unused variables 2003-10-27 00:41:09 +00:00
lukem af258d4e7c add mknod as a host tool 2003-10-27 00:35:37 +00:00
christos 1293ac6b7f remove buildcontext 2003-10-27 00:16:42 +00:00
christos 382a193ff0 eliminate buildcontext, and fix siginfo delivery. 2003-10-27 00:16:24 +00:00
lukem 6182ef67e2 be consistent using #if HAVE_STRUCT_STAT_ST_xxx (versus #ifdef) 2003-10-27 00:16:21 +00:00
christos 2c67a735b8 Provide the correct trap address to userland. 2003-10-27 00:15:24 +00:00
lukem 397ba902d6 enable ident strings for sources pulled in from libc 2003-10-27 00:14:59 +00:00
lukem 171d653219 Overhaul how `build.sh tools' are used:
*	Rename "config.h" to "nbtool_config.h" and
	HAVE_CONFIG_H to HAVE_NBTOOL_CONFIG_H.
	This makes in more obvious in the source when we're using
	tools/compat/config.h versus "standard autoconf" config.h

    *	Consistently move the inclusion of nbtool_config.h to before
    	<sys/cdefs.h> so that the former can provide __RCSID() (et al),
    	and there's no need to protect those macros any more.

These changes should make it easier to "tool-ify" a program by adding:
	#if HAVE_NBTOOL_CONFIG_H
	#include "nbtool_config.h"
	#endif
to the top of the source files (for the general case).
2003-10-27 00:12:41 +00:00
kleink 5976f8aa56 Retire local union { floating-type; struct ieee_prec; } in favor of those
available from <machine/ieee.h>
2003-10-27 00:05:46 +00:00
chs 44d9d27c84 don't print an uninitialized variable. 2003-10-26 23:42:57 +00:00
cl 7833f0afc9 now that MAKEDEV is generate, install it from ${.OBJDIR} 2003-10-26 23:34:29 +00:00
chris 25c741b8a6 Fix up some unitialised variables. 2003-10-26 23:11:15 +00:00
lukem 3d29fcd1f7 Fix previous:
*	Restore previous functionality which supported running postinstall
	with an extracted etc.tgz set as the "source" directory (i.e,
	/usr/src isn't available).  We do this by generating the MAKEDEV
	to ${SCRATCHDIR} if the real source is available, and leverage off
	the compare_dir() code that the rest of postinstall uses to support
	this functionality.

    *	Don't special case MACHINE_ARCH==arm; the MAKEDEV.awk should
	be doing that since etc/Makefile doesn't have similar special
	case code.  In any case, AFAICT the only MACHINE_ARCHs are
	"arm" and "armeb"; there is no "armel" in NetBSD.
2003-10-26 23:08:29 +00:00
fvdl 4784db999a Regen after clarifying IGPE1000 name. 2003-10-26 22:25:53 +00:00
fvdl 4589fc061f Make the name of the IGPE1000 a bit clearer. 2003-10-26 22:24:58 +00:00