Aren/NetBSD
1
0
Fork 0
Code Issues Pull Requests Packages Projects Releases Wiki Activity
175,431 Commits 606 Branches 0 Tags 3.1 GiB
a5a0d68555
Commit Graph

644 Commits

Author SHA1 Message Date
mrg
8679e33545 if LD32DIR is defined: 
- look in src/compat/lib/libc for libc_pic.
- define RTLD_ARCH_SUBDIR to "i386" or "sparc" for amd64 and sparc64
  builds, respectively
if RTLD_ARCH_SUBDIR is defined, add this path before
RTLD_DEFAULT_LIBRARY_PATH in the default search path.
 2008-10-26 07:11:54 +00:00
christos
3b55965a05 Never build this with PIE. 2008-10-16 14:36:20 +00:00
skrll
8c9056ada8 lib/39649: dlsym(3) does not follow dependencies 
Pull across code from FreeBSD to do a search of the passed object and it's
NEEDED objects (dependencies).

Reviewed by gimpy.
 2008-10-04 09:37:12 +00:00
macallan
982ae38f15 avoid usind __builtin_return_address(0) on PowerPC because it fails in 
Xorg's module loader.
Should fix PR37812
 2008-09-27 03:52:05 +00:00
matt
86d6d815ae Make sure to call dlerror() to clear any error messages after the relocations 
are done and there are no errors.  This will prevent spurious undefined weak
symbol errors being returned by dlerror().

Fixes PR bin/39240.
 2008-07-29 16:27:01 +00:00
skrll
dccd9ddf7f RCSID police 2008-07-24 06:51:58 +00:00
matt
38bdc8954f Refactor common code to _rtld_relocate_plt_object to i386 and arm so they 
act like the other versions.
In _rtld_bind, if the result is 0, call _rtld_die.
Initialize _rtld_sym_zero.st_value to -_rtld_objself.maprelocbase.  Now when
the symbol is resolved, st_value + maprelocbase will equal 0 and the above
check in _rtld_bind will fire and a call to NULL will be avoided.
 2008-07-24 04:39:25 +00:00
christos
d3b459e727 abort() if we are trying to use an undefined weak symbol from the plt. Remove 
check from relocate_plt_object() since we cannot return _rtld_sym_zero anymore.
Code from gimpy.
 2008-07-23 18:16:42 +00:00
christos
0d33a45b97 handle sym_zero like x86_64; other ports might need this too. 2008-07-22 22:27:07 +00:00
christos
56e895e29f deal specially with sym_zero, returned by a weak symbol that was not found, 
by retrieving it's name instead of printing random garbage.
 2008-07-21 20:19:28 +00:00
ad
156176be13 Fix error in previous. From nakayama@. 2008-06-05 00:03:20 +00:00
ad
53c5ea5da9 Don't bother unmapping the ELF header unless it overlaps the first load 
section.
 2008-06-03 19:32:32 +00:00
ad
9795e155ec __progname isn't const. 2008-06-03 19:22:07 +00:00
ad
36b49f84f3 Try read() on ld.so.conf. If the file is small, it avoids stat+mmap+munmap. 2008-06-03 18:36:59 +00:00
wiz
488200e6ea Remove trailing whitespace and superfluous .Pp. 2008-05-25 19:48:35 +00:00
martin
11a6dbe728 Convert TNF licenses to new 2 clause variant 2008-04-30 13:10:46 +00:00
martin
ce099b4099 Remove clause 3 and 4 from TNF licenses 2008-04-28 20:22:51 +00:00
yamt
8c93baa802 constify 2008-01-14 08:53:42 +00:00
yamt
8a58f724e1 tabify 2008-01-14 08:52:16 +00:00
christos
8fb14189dc make this compile again. 2007-12-29 01:44:03 +00:00
christos
716fde8fa0 simplify; mapbase == vaddrbase from the PaX author. 2007-12-28 19:36:21 +00:00
christos
8636ab52f6 Deal with ASLR PIE executables, by recognizing that phdr != phdr->v_addr 
for PT_PHDR headers when the program has been loader to a vaddr other than
then one specified (a randomized one), and modifying the relocation base
address appropriately (idea from elad)
 2007-12-26 21:14:41 +00:00
christos
8b0637156f No need for stack_protector.o if USE_FORT is no. 
XXX: compiling a binary with -Wl,--dynamic-linker,${NETBSDSRCDIR}/libexec/...
results in stack overflow. Investigate!
 2007-12-26 21:11:52 +00:00
christos
5ad1b72097 print the symbol we are seaching for [when debugging] 2007-12-26 21:10:03 +00:00
ad
558a5ffd0f rtld changes for PR bin/37347: 
- Apply patch from J.T. Conklin to execute .init/.fini functions in order.
- Support DF_1_INITFIRST and mark libc with DF_1_INITFIRST. Shared libs
  should be recording a dependency on libc, but it's too late to do that.

Ok christos@.
 2007-12-07 20:34:04 +00:00
christos
496f9be3e8 support %z 2007-11-24 18:32:26 +00:00
ad
71ac015ebf stack_protector.c has moved. 2007-11-13 15:34:51 +00:00
skrll
29353f93e2 Build a local copy of exit that doesn't call __cxa_finalize. It's not 
needed and drags in a lot of libc via free.
 2007-10-30 17:19:59 +00:00
christos
ed1a8fea8d build stack protector locally so that we don't call syslog_ss that results 
in 1/2 of libc dragged in.
 2007-10-29 20:36:31 +00:00
ad
bf4b000d16 - Don't use the names malloc/free/etc, since we can end up pulling in libc's 
malloc which is not what we want.
- Remove ancient memory compaction hack. Realloc on free memory is now an
  error again (from joerg@).

Ok christos@.
 2007-10-05 22:21:07 +00:00
christos
596c6ec5db provide xrealloc, and don't cast and use malloc and realloc directly. 2007-09-27 17:52:16 +00:00
christos
2cce1500a2 If we are using SSP, compile in a separate copy of the stack protector 
functions so that we don't end up calling syslog_ss() which drags in
a large portion of libc. Found by skrll...
 2007-09-27 17:51:40 +00:00
christos
6f875d32e2 Don't trust AT_SUN_EXECNAME, if we are set{u,g}id. 2007-09-21 13:57:09 +00:00
christos
6eac115581 delete debug printf 
make regress compile again.
 2007-09-20 19:42:53 +00:00
christos
40cfdb5e3a Use AT_SUN_EXECPATH instead of argv[0] for the executable name. This needs 
kernel support, which is not there yet, thus $ORIGIN is broken at this time.
This is ok, nothing uses it yet.
 2007-09-20 14:14:25 +00:00
cube
f7779133cd Don't remove the last component of filename before calling 
_rtld_digest_dynamic:  it is done in expand(), and doing it twice is once
too much.  This fixes $ORIGIN expansion for dependencies.
 2007-09-20 07:50:32 +00:00
tnn
ea0fc2f827 5 is not a valid version for .Nx, so change it to 5.0. 2007-09-13 21:32:14 +00:00
joerg
ad61ac15d2 Don't define bool, false and true. Use stdbool.h. 2007-09-02 09:57:16 +00:00
christos
2ffe4b875f Use an elf note to handle pax arguments. This is a temporary solution to 
avoid wasting OS flag bits. In the future we'll probably use fileassoc to
achieve this (once there is a way to make fileassoc persistent) or in the
shorter term libelf, so that we can add and remove the note on demand instead
of burning bits on each binary. Of course since this is a tool, this means
that we'll need to think about how to handle libelf...
 2007-06-24 20:35:36 +00:00
christos
3b320420fe fix a comment. 2007-05-18 21:46:39 +00:00
christos
04ee85ad50 Support Solaris-like $ORIGIN etc. expansions in paths. 2007-05-18 21:44:08 +00:00
scw
596a371d09 Remove a test for defined(__SH5__). 2007-04-08 10:02:35 +00:00
scw
5764a76889 Remove support for NetBSD/{,evb}sh5. 2007-04-08 09:35:21 +00:00
matt
549524f84f Deal with MKPICLIB != no. 2007-03-15 17:27:15 +00:00
matt
ee98d74a6b Remove MD COMBRELOC support since it's now MI. 
Enable COMBRELOC for all arches
 2007-02-23 01:17:11 +00:00
matt
364993dede Don't use cached results for plt lookups (nor save them). 2007-02-23 01:16:32 +00:00
matt
4d10e5e667 Make the COMBRELOC support generic. 2007-02-22 18:57:48 +00:00
martin
3cc1de3d7b Add COMBRELOC support 2007-02-15 19:42:13 +00:00
skrll
1fddd99f52 Provide COMBRELOC support for amd64 in answer to PR/35589. 
Some statistics:

base + COMBRELOC
$ time (for i in `seq 100`;do noatun --help>/dev/null;done)
(; for i in `seq 100`; do; noatun --help > /dev/null; done; )  148.64s
user 4.82s system 99% cpu 2:33.93 total

base + DF caching:
$ time (for i in `seq 100`;do noatun --help>/dev/null;done)
(; for i in `seq 100`; do; noatun --help > /dev/null; done; )  151.15s
user 5.53s system 99% cpu 2:37.23 total

base:
$ time (for i in `seq 100`;do noatun --help>/dev/null;done)
(; for i in `seq 100`; do; noatun --help > /dev/null; done; )  492.36s user
5.34s system 99% cpu 8:19.17 total
 2007-02-15 15:44:28 +00:00
wiz
09cb1d6f1c s/existance/existence/, from Zafer. 2006-11-24 22:52:16 +00:00