summaryrefslogtreecommitdiffstats
path: root/lib/csu (follow)
Commit message (Collapse)AuthorAgeFilesLines
* Define both _start and __start in crt0, all of the other architectures do.drahn2003-08-252-3/+9
|
* ansideraadt2003-06-271-9/+5
|
* boring protosderaadt2003-06-262-10/+10
|
* more 3/4 cleanups for lazy developersderaadt2003-06-041-7/+2
|
* i sent you the damn working diff, didn't i?mickey2003-05-301-5/+7
|
* Fix dependancy building for ELF startup code. Changed Makefilesdrahn2003-05-306-96/+157
| | | | to build startup code the same on all (ELF) archs. hppa ok mickey@
* i386 startup code for ELF. based on newer version of NetBSD code.drahn2003-04-172-84/+98
|
* add a newline at end of file. gcc3 rightfully complains if the filerohee2003-03-312-4/+4
| | | | | | is not newline terminated. From Thorsten Glaser
* duplicate words and spelling fixes in commentsdavid2003-03-101-3/+3
| | | | ok miod@
* copy basename of argv[0] to bss for __progname, so that large stack smashderaadt2003-02-2810-20/+80
| | | | does not make propolice reporting ineffective; millert miod ok
* Switch to a C version of powerpc crt0 file. easier to read. movesdrahn2003-02-263-90/+103
| | | | _progname storage into data instead of stack.
* extern the eprol and etext; better eprol definitionmickey2003-01-161-3/+4
|
* build crtbegin/crtend as it is ment for elfmickey2002-11-102-22/+59
|
* uncommit, since it breaks macppc libc. millert and miod have now said theyderaadt2002-09-174-19/+19
| | | | | | did NOT approve those for commit. why did mickey feel he was ok to go commiting a set of diffs which had not been passed around and tested by people? we don't know. mickey, have you got something to say for yourself?
* use a different name for __syscall(); millert@ miod@ niklas@ okmickey2002-09-174-19/+19
|
* Adapt sparc crt* to how things are done on our other ELF architectures.art2002-08-112-145/+105
| | | | Some parts of crt0.c are from NetBSD.
* Remove some dead ifdefs to make the code more readable.art2002-07-223-79/+4
|
* Fix an oversight in revision 1.9. ld_entry was set in one more spot.art2002-07-141-2/+2
|
* We need to map the data segment of ld.so with PROT_EXEC for the GOT.art2002-07-111-3/+7
| | | | | | This is in preparation for actually honoring exec protection in sparc pmap. NOTICE! You really want to rebuild the whole userland with this change before the kernel fix goes in. Otherwise all shared binaries will dump core.
* one more level of indirection for dlopen structure.espie2002-06-051-14/+14
| | | | | | | | Allows dlopen to be used much earlier, e.g., in dynamic constructors. From NetBSD. okay art@
* Manual cleanup of remaining userland __P use (excluding packages maintained outside the tree)millert2002-02-171-2/+2
|
* Part one of userland __P removal. Done with a simple regexp with some minor hand editing to make comments line up correctly. Another pass is forthcoming that handles the cases that could not be done automatically.millert2002-02-1618-86/+86
|
* fix some protosmickey2002-02-021-3/+5
|
* add sparc64; op2@squish.orgjason2001-09-111-2/+2
|
* I don't even want to know how I could screw this up so bad and noone noticed.art2001-09-021-2/+2
|
* build crt0, crtbegin and friends.art2001-08-291-0/+51
|
* crt0. Almost like on alpha, except that sparc64 needs an assembler stubart2001-08-291-0/+111
| | | | | for the moment. (why make the ABI so complicated?) The assembler is inspired by crt0 in NetBSD.
* remove unused bits and pieces.brad2001-08-193-162/+2
| | | | | -- Ok'd by art@
* make obj in hppa alwaysmickey2001-08-191-2/+2
|
* either way this is what i'm using nowmickey2001-08-192-0/+130
|
* #(endif|else) foo is incorrect, make it #endif /* foo */heko2001-08-124-12/+12
| | | | deraadt@ ok
* Wallderaadt2001-07-181-3/+4
|
* remove mips leftovers; ok bradpvalchev2001-07-115-259/+2
|
* Put m88k-related files in the correctly named directory.miod2001-06-263-5/+5
|
* proto for dlerror() changedderaadt2001-06-011-2/+2
|
* fix the alpha after that last diff which was obviously not testedderaadt2001-05-291-3/+3
|
* Commonize csu code for elf systems, powerpc now no longer has it's owndrahn2001-05-285-28/+18
| | | | | | | | | | | | | | versions of these files. Fixed a bug in ld.so in this, instead of scheduling the fini of each of the shared libraries with atexit. schedule a function of ld.so itself and it will walk all of the open libraries when the program exits. otherwise a shared library could be dl_open()ed and then dl_close()d and then it would not be mapped for the atexit processing. TODO: What if atexit is not found (process did not link against libc?) Do shared libraries that are dl_closed have their global destructors run?
* MAP_COPY -> MAP_PRIVATEart2001-05-111-4/+4
|
* Use "realinstall" to allow the framework to do other work if needed;tholo2001-03-2410-19/+19
| | | | path of least surprise. Ok millert@
* Remove even more dead code.art2001-02-031-28/+3
|
* oops.art2001-02-032-8/+6
|
* Schedule running of __fini in __init, not __start.art2001-02-033-5/+6
|
* An OpenBSD tag doesn't hurt.art2001-02-032-0/+2
|
* Add an OS note identifying OpenBSD binaries.art2001-02-034-2/+46
| | | | This appears to be the standard way to do it.
* Decruftification.art2001-02-032-16/+2
|
* Use crtbegin and crtend from common_elf.art2001-02-033-129/+4
|
* Create a common_elf directory with crtbegin and crtend that canart2001-02-034-0/+250
| | | | be shared between archs.
* Don't define ECOFF_COMPAT. We're ELF.art2001-02-031-5/+1
|
* deconfuse include path. (XXX - this for my future hacks).art2001-01-311-2/+2
|
* This is not how we'll do dynamic libraries on alpha anyway.art2001-01-252-102/+17
| | | | Zap dead code.