Commit Graph

22 Commits

Author SHA1 Message Date
manu 4d607adf4f Load __OBJC and ____CGSERVER sections of Mach-O binaries as __TEXT. 2002-12-11 19:28:41 +00:00
manu 39b522c107 back out the previous change, which is useless. Darwin loads the libraries
that are required by the binary, not the libraries required by the libraries
required by the binary.

Hopefully, binaries should load again on i386.
2002-11-29 15:49:09 +00:00
manu b36d0c1bf6 Maitain a chainedlist of already loaded Mach-O objects, to avoid loading
the same file multiple times because of recursive loading (ie: libx require
liby and libz and liby require libz, so libz would be loaded twice)

This is probably suboptimal, but it enable /bin/sh to load on the PowerPC,
so it's a good interim solution until we figure precisely how things should
work.

I'm not sure whether this makes the excessive recursive check useless or not.
2002-11-29 11:31:11 +00:00
manu 589ff87f68 Rewrite the excessive recursive loading protection by actually counting the
recursions instead of the total function calls. We limit to 6 recursion,
which is what Darwin does.
2002-11-24 21:59:43 +00:00
manu 0e734a6755 Libraries do not use relative addresses, they are absolute and should be
treated as such if we want the same mapping as in Darwin.

While we are there KNFify function names
2002-11-22 23:09:46 +00:00
manu f528c56f39 Check for excessive recursive Mach-O loading 2002-11-21 22:30:32 +00:00
manu ae6b5562b1 Remove a debug message that has been committed by mistake. 2002-11-21 22:01:45 +00:00
manu af59b63bbd We now have the exact stack initial stack layout of Darwin:
macho_hdr, argc, *argv, NULL, *envp, NULL, progname, NULL,
*progname, **argv, **envp

Where progname is a pointer to the program name as given in the first
argument to execve(), and macho_hdr a pointer to the Mach-O header at
the beginning of the executable file.
2002-11-21 19:53:40 +00:00
christos e6535bf628 XXX: be32toh on powerpc does not cast to int32_t so passing u_long to
be32toh produces an unsigned long result, causing a printf argument
mismatch. This is the wrong fix, but I am not going to change the
powerpc macros; fix the powerpc macros and revert my change.
2002-10-31 02:40:41 +00:00
manu 293419f314 int format long arg problems 2002-10-30 23:08:00 +00:00
christos 74e8fc3424 fix fat binary architecture choosing code. mach binaries now execute as
poorly as before the x86 MP merge.
2002-10-30 15:21:01 +00:00
christos dc5aff13e6 make this compile again. 2002-10-30 15:04:47 +00:00
manu 9b4f3b0464 Introduce an array of supported CPU types by a given arch for Mach-O 2002-10-29 22:22:30 +00:00
manu b57515f2c4 If the magic number is 0xfeedface instead of 0xcafebabe, this means that
the executable is not fat. The binary starts with an object header instead
of a mach fat header.
2002-10-29 19:28:19 +00:00
chs 993948e989 count executable image pages as executable for vm-usage purposes.
also, always do the VTEXT vs. v_writecount mutual exclusion
(which we previously skipped if the text or data segment was empty).
2002-10-05 22:34:02 +00:00
provos 0f09ed48a5 remove trailing \n in panic(). approved perry. 2002-09-27 15:35:29 +00:00
lukem adc783d537 add RCSIDs 2001-11-12 15:25:01 +00:00
christos 5c7505a94f fix compile typo. 2001-11-10 20:49:52 +00:00
thorpej e8ee04475d - Add a new vnode flag VEXECMAP, which indicates that a vnode has
executable mappings.  Stop overloading VTEXT for this purpose (VTEXT
  also has another meaning).
- Rename vn_marktext() to vn_markexec(), and use it when executable
  mappings of a vnode are established.
- In places where we want to set VTEXT, set it in v_flag directly, rather
  than making a function call to do this (it no longer makes sense to
  use a function call, since we no longer overload VTEXT with VEXECMAP's
  meaning).

VEXECMAP suggested by Chuq Silvers.
2001-10-30 15:32:01 +00:00
christos f7cb99f9c6 make a copy of the epp->ep_path, because it is too late to use it when
copyargs is called since we've swapped vmspaces.
2001-10-28 04:46:42 +00:00
christos 66b3b42040 use be32toh instead of bswap32
add a macro MACHO_MACHDEP_CASES instead of ifdef'ing cpu types.
thanks jason!
2001-07-14 03:05:31 +00:00
christos b6639a0e83 new exec module for MACH-O fat binaries
XXX: shared libraries are not handled properly yet.
2001-07-14 02:09:41 +00:00