Commit Graph

3444 Commits

Author SHA1 Message Date
tron 2ece7637b1 Add "_eh.c" to list of files which are supposed to get cleaned. 2002-01-02 09:17:48 +00:00
thorpej 2135cbaadd Oops, this is NOT a C++ program. 2002-01-01 05:16:18 +00:00
thorpej d68ee9d8e3 Per a suggestion from matthew green, PROG_CXX takes the place of
PROG for C++ programs.
2002-01-01 01:38:25 +00:00
thorpej a156b86123 Set PROG_CXX since these are C++ programs. 2002-01-01 00:27:33 +00:00
thorpej 89b1a9ca73 Introduce 3 new MK* variables that have effect in the USE_NEW_TOOLCHAIN
case:

	MKBFD	If set to "no", disables building of libbfd, libiberty,
		and all things that depend on them (binutils/gas/ld, gdb,
		dbsym, mdsetimage).

	MKGDB	If set to "no", disables bulding of gdb.

	MKGCC	If set to "no", disables building of gcc and the
		gcc-related libraries (libg2c, libgcc, libobjc, libstdc++).

These are useful for building platforms for which either of the following
situations are true:

	(1) You have no userland from which to run toolchain2netbsd
	    in order to build the appropriate toolchain build framework.

	(2) The platform which you are building requires a newer set
	    of tools than are currently in the tree (e.g. x86-64, ia64).
2001-12-31 23:04:11 +00:00
mrg d750ba5f81 this is really libstdc++ 2.8.0 2001-12-30 17:21:19 +00:00
tv f14c689431 Fix %{p*:} botch, as described in toolchain/15024. (What specs need is an
if/then/else type construct....)
2001-12-29 20:24:37 +00:00
lukem efcc9a4c9d * Add user-controlled mk.conf variables
- SHLIBDIR	Location to install shared libraries if ${USE_SHLIBDIR}
			is "yes".  Defaults to "/usr/lib".

	- USE_SHLIBDIR	If "yes", install shared libraries in ${SHLIBDIR}
			instead of ${LIBDIR}.  Defaults to "no".
			Sets ${_LIBSODIR} to the appropriate value.
			This may be set by individual Makefiles as well.

	- SHLINKDIR	Location of shared linker.  Defaults to "/usr/libexec".
			If != "/usr/libexec", change the dynamic-linker
			encoded in shared programs

* Set USE_SHLIBDIR for libraries used by /bin and /sbin:
	libc libcrypt libcrypto libedit libipsec libkvm libm libmi387
	libtermcap libutil libz

* If ${_LIBSODIR} != ${LIBDIR}, add symlinks from ${LIBDIR}/${LIB}.so*
  to ${_LIBSODIR}/${LIB}.so* for compatibility.

* Always install /sbin/init statically (for now)


The net effect of these changes depends on how the variables are set:

  1.)	If nothing is set or changed, there is no change from the
	current behaviour:
		- Static /bin, /sbin, and bits of /usr/*
		- Dynamic rest
		- Shared linker is /usr/libexec/ld*so

  2.)	If the following make variables are set:
		LDSTATIC=
		SHLINKDIR=/lib
		SHLIBDIR=/lib
	Then the behaviour becomes:
		- Dynamic tools
		- .so libraries used by /bin and /sbin are installed to /lib,
		  with symlinks from /usr/lib/lib*so to -> /lib/lib*so
		  where appropriate
		- Shared linker is /lib/ld*so

  3.)	As per 2.), but add the following variable:
		USE_SHLIBDIR=yes
	This forces all .so's to be instaleld in /lib (with compat
	symlinks), not just those tagged by their Makefiles to be.
	Again, compat symlinks are installed
2001-12-28 01:32:37 +00:00
mycroft 913449a520 Disable the atexit(_IO_cleanup) for two good reasons:
1) There is already a destructor that does this.  The atexit() is superfluous.
2) If libstdc++ is loaded dynamically by dlopen() and then is removed by
   dlclose(), the pointer is no longer valid and the program would core dump
   at exit() time.  (This caused xmms to core dump at exit when xmms-sid was
   installed, for example.)
NOTE: THIS NEEDS TO BE APPLIED TO THE `TOOLCHAIN' VERSION.
2001-12-24 16:38:19 +00:00
matt 46cf0c1543 Incorporate some gcc3.0 vax.md changes 2001-12-24 01:56:55 +00:00
matt c9073b0a26 Expand VAX ELF relocs into new style (with comments). Nuke OSABI shit. 2001-12-24 01:55:32 +00:00
thorpej 0146257365 Set NOMAN correctly. From Andrew Brown. 2001-12-21 08:06:15 +00:00
thorpej 3778dfc4fd Build as "protoize" and "unprotoize" respectively, not "gcc" (!!). 2001-12-21 08:03:35 +00:00
thorpej f7fd432381 Fixes to alpha_next_pc() (for software single-step):
* jump-format jumps: the target is in RB, so use the RB bits in
  the insn, not the RA bits.
* branch-format displacements are 21 bits; extract all of them, and
  sign-extend correctly.
* For conditional branches, we need to determine whether or not
  the branch is actually taken.
2001-12-21 07:58:47 +00:00
matt fb5f3f3500 Minimal changes/files to get gdb for vax building under the new toolchain. 2001-12-21 06:36:10 +00:00
thorpej 4644cdb471 Implement software single stepping, and enable it for NetBSD/alpha. 2001-12-21 02:57:38 +00:00
thorpej 38b00971f1 Update for Alpha emulation change. 2001-12-21 02:03:20 +00:00
thorpej 1c899ee8d9 Update for changes to alphanbsd-nat.c. 2001-12-21 01:57:42 +00:00
thorpej 011356e8eb Major overhaul, plus support ELF core files. 2001-12-21 01:57:06 +00:00
yamt 3dd7e76ccf fix typo. (.o in SRCS) 2001-12-20 22:18:50 +00:00
thorpej d103096c14 Fixup VERSION. 2001-12-20 18:26:30 +00:00
scw 568cb5d556 Revert the last change following yesterday's commit to ../netbsd.h
(Hi Jason!)
2001-12-20 18:14:35 +00:00
rearnsha 2bc7fae481 Avoid generating code sequences that might cause alignment faults. 2001-12-20 00:50:27 +00:00
thorpej c3af5af42b Redo previous differently: Put the ASM_DECLARE_OBJECT_NAME
and ASM_FINISH_DECLARE_OBJECT macros in <alpha/netbsd.h>,
and remove ASM_DECLARE_FUNCTION_NAME, ASM_DECLARE_OBJECT_NAME,
ASM_FINISH_DECLARE_OBJECT, and ASM_DECLARE_FUNCTION_SIZE from
<netbsd.h>.
2001-12-20 00:18:01 +00:00
tv 4c9a2ef7d3 * Add protoize and unprotoize. (Closes PR 6894.)
* Regen files with proper OS names and version numbers.

* Clean up toolchain2netbsd somewhat, to get it ready to be cross-host
  compatible (more work to be done here, but it's getting closer).

* Add framework for gdbreplay and gdbserver, but hold off on enabling these
  by default until low-nbsd.c is verified to work everywhere.
2001-12-19 23:54:16 +00:00
thorpej 19241b97b5 Recent change to <netbsd.h> broke ASM_DECLARE_FUNCTION_NAME. Re-define
it here to what it should be.
2001-12-19 23:30:53 +00:00
rearnsha 0e644c95f7 Implement ATPCS stack-alignment rules, and enable on arm-netbsdelf
configuration.
2001-12-19 23:14:50 +00:00
tv 14691f972b Remove COPTS+=-g. This should be set by the builder, not this Makefile. 2001-12-19 21:55:48 +00:00
tv 0a1ba8ecaa When NETBSD_NATIVE, use the appropriate version of libgcc for the compile
mode:  libgcc_pic.a for -shared, and libgcc_p.a for -p[g].
2001-12-19 21:38:24 +00:00
tv a3aeedcdb7 Tag all version strings with "nb1" to make FSF/binutils/gcc/gdb people
happy.  (This flags the version as deviating from the official FSF
distribution inside the NetBSD source tree.)
2001-12-19 21:32:49 +00:00
tv 7fb95bfb35 Remove iostdio.h from the list of system-installed headers; intended to
address parts of PR toolchain/14896.  This header file is nonstandard
(and doesn't even exist in gcc 3.0); an out-of-the-box gcc build also
doesn't provide the missing functions.  So just drop the .h completely.
2001-12-19 20:52:11 +00:00
scw 37f114189c G/C some #define's which are already in ../netbsd.h.
Gets libgcc compiling again.
2001-12-19 18:14:20 +00:00
thorpej 06811a827d Go back to the nop/unop pattern, and pick up rev. 1.35 from the
master FSF sources (rth):

	* config/tc-alpha.c (alpha_handle_align): Encode unop with RB as $sp.
2001-12-18 00:33:35 +00:00
thorpej cc4beed398 From the master FSF sources, rev. 1.7 (rth):
* alpha-opc.c (unop): Encode with RB as $sp.
2001-12-18 00:31:52 +00:00
thorpej c7af981281 Make a note about what "nop" and "unop" actually are. 2001-12-17 01:51:44 +00:00
thorpej e93b1005a7 Apply rev 1.4 of gnu/dist/gcc/config/sh/sh.h:
fix REGISTER_MOVE_COST bug (gcc original rev. 1.43):

> *sh.h (REGISTER_MOVE_COST): Change PR_REG to PR_REGS
2001-12-16 18:56:45 +00:00
thorpej 3274936f9b Apply rev 1.3 of gnu/dist/gcc/config/sh/sh.h (fix ASM_OUTPUT_REG_PUSH
and ASM_OUTPUT_REG_POP):

fix two bugs:
	- cc1 dumps core when "-pg" is used
	- gas produces a syntax error
2001-12-16 18:54:10 +00:00
thorpej d63c80c2c6 Apply rev. 1.2 of gnu/dist/gcc/config/sh/sh.h:
set DWARF_LINE_MIN_INSTR_LENGTH to 2 because the length of instruction
is fixed to 16bits.
2001-12-16 18:50:16 +00:00
thorpej c8a285e813 Apply rev. 1.7 of gnu/dist/config/sh/sh.c:
apply gcc original rev. 1.59:

 > * config/sh/sh.c (prepare_scc_operands): Apply force_reg to
 > sh_compare_op1 when the mode is DImode and sh_compare_op1 isn't
 > const0_rtx.
2001-12-16 18:47:47 +00:00
thorpej ff7287ad08 Apply change from rev 1.5 of gnu/dist/gcc/config/sh.c:
fix output_ieee_ccmpeq() bug:

- 2nd argument is not rtx but rtx *
- return value of output_ieee_ccmpeq is not void but char *
2001-12-16 18:38:19 +00:00
scw f87b859f0d After a discussion with Richard Henderson (rth at redhat dot com), add
the patch I posted to tech-toolchain@netbsd.org as a workaround for the
GOT bug whereby the callee-saved register %a5 can be trashed in certain
optimised -fPIC functions.

This may yet need to be fixed a different way, pending input from
Andreas Schwab on his change in revison 1.33 of this file in the
FSF source...

In the meantime, I've built and installed several NetBSD shared libraries
(including libc) with my change and found no problems.
2001-12-16 18:28:15 +00:00
thorpej 037cacf7e2 Define __NO_LEADING_UNDERSCORES__, since this sh-elf toolchain doesn't
use them.
2001-12-16 18:09:46 +00:00
thorpej 75acdbe31a Undo a local change inherited from our local binutils 2.9 that
is not needed in binutils 2.11.
2001-12-16 17:50:56 +00:00
thorpej f6e3c2a3da Make this compile. Still needs work. 2001-12-16 07:26:22 +00:00
thorpej 583334b6ca Avoid infinite #include recursion. 2001-12-16 07:19:42 +00:00
thorpej c17876b1d2 Use the elf64alpha_nbsd linker emulation. 2001-12-16 04:26:17 +00:00
thorpej 7a362db4f3 After discussing the entry point issue w/ rth@redhat.com, add an
elf64alpha_nbsd linker emulation that sets the entry point correctly
for NetBSD.
2001-12-16 04:06:51 +00:00
thorpej e3e5021be2 We need the versions of ASM_DECLARE_FUNCTION_NAME and
ASM_DECLARE_FUNCTION_SIZE from <alpha/alpha.h>, but <netbsd.h>
re-defines them.  So, re-define them again here to what
we need.
2001-12-16 04:02:48 +00:00
thorpej 72a263b8f7 Catch up with other Alpha-ELF changes. This will get cleaned up
even more somtime soon.
2001-12-16 03:09:19 +00:00
thorpej 80ed7d6a42 Clean up a bunch of the ELF support (looks like there was some sort
of merge error), and make the function/object type/size macros work
with the Alpha.
2001-12-16 03:07:59 +00:00