| Age | Commit message (Collapse) | Author |
|---|
|
Now all functions and vaiables can be made static to reduce relocs.
Basicly, all variables should be merged into one struct variable
and then export that struct. Exported functions should be exported
as functions pointers in the above struct. This will reduce the
number of relocs inside ldso to a handful.
Add -Bsymbolic to linker for ldso. This will resolve all functions
(all JMP_SLOT goes away) and the remaining relocs are transformed into
RELATIVE relocs.
|
|
|
|
misread this code and removed it, sorry.
I printed the values of _dl_brkp and _dl_envp in ldso and
got this when executing:
ls:
__curbrk: 0x300c16e0
__environ: 0x1008372c
ldd:
__curbrk: 0x300776e0
__environ: 0x3007759c
why is__environ in the app when executing ls but not ldd?
|
|
|
|
|
|
|
|
the whole lot in a single pass.
-Erik
|
|
|
|
This removes some crap in libdl.c(and future libs which needs
to access ldso functions).
What do you think?
|
|
Fixed a bug in libdl.c where the RTLD_GLOBAL was assigned
wronly.
|
|
only the global GOT entry relocations for the loaded shared libs were being
handled. This made applications segfault when using un-fixedup symbols. After
many hours debugging to track down the problem, this patch makes the mips port
happy again.
-Erik
|
|
"I fixed the _dl_parse_copy_information in the same way than for the
powerpc and it works fine for me.
You may luck at the patch for the powerpc/elfinterp.c, where I change
the paramter of a dl_dprintf statement. Now we use the same
parameters than for the relocation copy."
Arch mantainers, please do the same.
When all arches has been fixed, we can remove of _dl_parse_copy_information,
it is just a waste of CPU cycles.
|
|
Hi Manuel & Erik
I think I know why MIPS is broken. _dl_perform_mips_global_got_relocations() is
broken. It will due to my latest changes reloctate ldso. This
function needs to die and its job should be done inside _dl_parse_relocation_information().
It is mostly a copy and paste job,
Also PERFORM_BOOTSTRAP_GOT and PERFORM_BOOTSTRAP_RELOCATION should be fixed, they
use symbols which aren't passed as arguments.
Jocke
|
|
Remove unused function parameter in ldso.
Remove echo option '-e' since it is not supported on Solaris.
|
|
|
|
to use '_dl_pagesize'
-Erik
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
I finally tracked down the problem with SH4 shared library loading -
in file ldso/ldso/sh/dl-startup.h, the line
#define DL_BOOT(X) static void __attribute_used__ _dl_boot2 (X)
should be
#define DL_BOOT(X) static void* __attribute_used__ _dl_boot2 (X)
otherwise the application code will never be called and _dl_boot will
just return causing a segmentation fault.
Carl
|
|
"There's a bug in the SH asm, it calls into 'C' code without preserving
all the important scratch registers. I have been meaning to double
check the changes as I think more registers need to be saved, then I
was going to send it back to uClibc. From the sound of it this isn't
your problem, but you never know.
Attached is the patch as it stands for the loader on SH."
Glibc appears to do the same.
|
|
is needed if ldso should use libcs malloc whenever possible.
Fix RTLD_LAZY propagation to RTLD_NOW relocation when requested by
libdl.
|
|
|
|
as the _dl_pagesize global, so do not alias the global.
-Erik
|
|
-Erik
|
|
Don't' use atexit in libdl.
Make destrutor in libdl actually run and to dlclose all
dlopened libs.
Remove dlclosed libs from the _dl_symbol_tables list.
|
|
Other arches may also benefit from this iff it can do
unaligned stores.
|
|
|
|
_dl_strchr,_dl_strrchr,_dl_strstr,_dl_memcmp:
Optimize for archs which can do pre increment/decrement and load/store
in one instruction.
|
|
handle "" strings and optimze it.
_dl_simple_ltoa,_dl_simple_ltoahex:
Optimize for archs which can do pre increment/decrement and load/store
in one instruction.
|
|
|
|
i was sticking some libraries in non-standard (i.e. not-hardcoded) paths and
then updating ld.so.conf accordingly ... oddly though, i couldnt get binaries
that linked against those libraries to run ... `ldconfig -v` showed the
required libraries being detected/cached, so something seemed wrong ...
turns out the libraries were being tagged as LIB_ELF_LIBC0 in the cache and
the loader currently only accepts LIB_ELF_LIBC{5,6} ... since readsoname.c in
ldconfig defines needed_tab with the mapping 'libc.so.0' -> LIB_ELF_LIBC0 it
seems like the loader should accept this type too
find attached a small patch which adds LIB_ELF_LIBC0 along side
LIB_ELF_LIBC{5,6} as an accepted lib type in the loader when dealing with the
ld.so.cache file
|
|
in libdl, the XXFLAGS is set incorrectly if DODEBUG=y ... it has 'I.' added to
it instead of '-I.' ... this of course pisses of gcc :)
find attached a patch that cleans up the XXFLAGS setting so as to minimize
this kind of thing ... when asked, mjn3 said it was a kosher idea ;)
and later writes:
hmm it seems ldso also utilizes USE_CACHE in the code (but LIBRARY_CACHE never
propogates down to the Makefiles)
so the fix is to add $(LIBRARY_CACHE) while setting $(XXFLAGS)
Applied with minor fixes and also fixed up ldso/Makefile to match
libdl/Makefile.
|
|
However RTLD_LOCAL still doesn't work. Everything is RTLD_GLOBAL.
|
|
PPC32, SPARC32/64 and S390 includes the PLT in its RELA size. This caused ldso
to always do unlazy relocation of the JMPRELs. This patch fixes it.
|
|
This patch makes -fpic work for PCC and optimzes the relcation by moving the cache
flushing stuff to JMP relocs only. Actually PPCs ldso can only handle small
GOT tables(<=8192 entries)anyhow, so it makes little sense to compile PPC with -fPIC.
libuClibc shrunk from 340724 to 330780 bytes with -fpic.
|
|
Some utilities, such as valgrind, have a legitimate reason to know the address
of the current brk. Since we know such utils will peek under our skirt, we
might as well give them what they expect and not use a gratuitously different
symbol name.
-Erik
|
|
rather than _dl_malloc, fprintf rather than _dl_printf, etc.
-Erik
|
|
|
|
|
|
|
|
tree. Bad boy, No doughnut.
-Erik
|
|
aligned pointer that may be aligned up to page_size. Also add _dl_free,
|
|
Hi yet again :)
in dl-startup.c when performing boot strap relocation the following test
exists to make sure that only "_dl_" symbols are relocated:
/* We only do a partial dynamic linking right now. The user
is not supposed to define any symbols that start with a
'_dl', so we can do this with confidence. */
if (!symname || !_dl_symbol(symname)) {
continue;
}
However on PPC(and the other archs as well I suspect) all symbols are
"_dl_" symbols so the test is never true. The test can be removed and the
whole loop simplified(smaller). This also makes it possible to
simplify elfinterp.c
This remove the scanning of ldso.so relocs, making relocation faster.
I have tested this on PPC and it works well.
Do you think this optimization will work for the other arches as well?
I can't see why not.
Jocke
* Tested on x86, arm, mipsel, and powerpc by Erik and works nicely
-Erik
|
|
only occur when shared libs are not compiled with -fPIC
|
