The force_first parameter was ineffective because the dlclose'd
object was not necessarily the first in the maps array. Also
enable force_first handling unconditionally, regardless of namespace.
The initial object in a namespace should be destructed first, too.
The _dl_sort_maps_dfs function had early returns for relocation
dependency processing which broke force_first handling, too, and
this is fixed in this change as well.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
The open_path stops if a relative path in search path contains a
component that is a non directory (for instance, if the component
is an existing file).
For instance:
$ cat > lib.c <<EOF
> void foo (void) {}
> EOF
$ gcc -shared -fPIC -o lib.so lib.c
$ cat > main.c <<EOF
extern void foo ();
int main () { foo (); return 0; }
EOF
$ gcc -o main main.c lib.so
$ LD_LIBRARY_PATH=. ./main
$ LD_LIBRARY_PATH=non-existing/path:. ./main
$ LD_LIBRARY_PATH=$(pwd)/main:. ./main
$ LD_LIBRARY_PATH=./main:. ./main
./main: error while loading shared libraries: lib.so: cannot open shared object file: No such file or directory
The invalid './main' should be ignored as a non-existent one,
instead as a valid but non accessible file.
Absolute paths do not trigger this issue because their status are
initialized as 'unknown' and open_path check if this is a directory.
Checked on x86_64-linux-gnu.
Reviewed-by: Szabolcs Nagy <szabolcs.nagy@arm.com>
The PR_SET_VMA_ANON_NAME support is only enabled through a configurable
kernel switch, mainly because assigning a name to a
anonymous virtual memory area might prevent that area from being
merged with adjacent virtual memory areas.
For instance, with the following code:
void *p1 = mmap (NULL,
1024 * 4096,
PROT_READ | PROT_WRITE,
MAP_PRIVATE | MAP_ANONYMOUS,
-1,
0);
void *p2 = mmap (p1 + (1024 * 4096),
1024 * 4096,
PROT_READ | PROT_WRITE,
MAP_PRIVATE | MAP_ANONYMOUS,
-1,
0);
The kernel will potentially merge both mappings resulting in only one
segment of size 0x800000. If the segment is names with
PR_SET_VMA_ANON_NAME with different names, it results in two mappings.
Although this will unlikely be an issue for pthread stacks and malloc
arenas (since for pthread stacks the guard page will result in
a PROT_NONE segment, similar to the alignment requirement for the arena
block), it still might prevent the mmap memory allocated for detail
malloc.
There is also another potential scalability issue, where the prctl
requires
to take the mmap global lock which is still not fully fixed in Linux
[1] (for pthread stacks and arenas, it is mitigated by the stack
cached and the arena reuse).
So this patch disables anonymous mapping annotations as default and
add a new tunable, glibc.mem.decorate_maps, can be used to enable
it.
[1] https://lwn.net/Articles/906852/
Checked on x86_64-linux-gnu and aarch64-linux-gnu.
Reviewed-by: DJ Delorie <dj@redhat.com>
Add anonymous mmap annotations on loader malloc, malloc when it
allocates memory with mmap, and on malloc arena. The /proc/self/maps
will now print:
[anon: glibc: malloc arena]
[anon: glibc: malloc]
[anon: glibc: loader malloc]
On arena allocation, glibc annotates only the read/write mapping.
Checked on x86_64-linux-gnu and aarch64-linux-gnu.
Reviewed-by: DJ Delorie <dj@redhat.com>
Linux 4.5 removed thread stack annotations due to the complexity of
computing them [1], and Linux added PR_SET_VMA_ANON_NAME on 5.17
as a way to name anonymous virtual memory areas.
This patch adds decoration on the stack created and used by
pthread_create, for glibc crated thread stack the /proc/self/maps will
now show:
[anon: glibc: pthread stack: <tid>]
And for user-provided stacks:
[anon: glibc: pthread user stack: <tid>]
The guard page is not decorated, and the mapping name is cleared when
the thread finishes its execution (so the cached stack does not have any
name associated).
Checked on x86_64-linux-gnu aarch64 aarch64-linux-gnu.
[1] 65376df582
Co-authored-by: Ian Rogers <irogers@google.com>
Reviewed-by: DJ Delorie <dj@redhat.com>
It adds NT_X86_SHSTK (2fab02b25ae7cf5), NT_RISCV_CSR/NT_RISCV_VECTOR
(9300f00439743c4), and NT_LOONGARCH_HW_BREAK/NT_LOONGARCH_HW_WATCH
(1a69f7a161a78ae).
All the crypt related functions, cryptographic algorithms, and
make requirements are removed, with only the exception of md5
implementation which is moved to locale folder since it is
required by localedef for integrity protection (libc's
locale-reading code does not check these, but localedef does
generate them).
Besides thec code itself, both internal documentation and the
manual is also adjusted. This allows to remove both --enable-crypt
and --enable-nss-crypt configure options.
Checked with a build for all affected ABIs.
Co-authored-by: Zack Weinberg <zack@owlfolio.org>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
This avoids crashes due to partially written files, after a package
update is interrupted.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
This reverts commit 6985865bc3.
Reason for revert:
The commit changes the order of ELF destructor calls too much relative
to what applications expect or can handle. In particular, during
process exit and _dl_fini, after the revert commit, we no longer call
the destructors of the main program first; that only happens after
some dlopen'ed objects have been destructed. This robs applications
of an opportunity to influence destructor order by calling dlclose
explicitly from the main program's ELF destructors. A couple of
different approaches involving reverse constructor order were tried,
and none of them worked really well. It seems we need to keep the
dependency sorting in _dl_fini.
There is also an ambiguity regarding nested dlopen calls from ELF
constructors: Should those destructors run before or after the object
that called dlopen? Commit 6985865bc3 used reverse order
of the start of ELF constructor calls for destructors, but arguably
using completion of constructors is more correct. However, that alone
is not sufficient to address application compatibility issues (it
does not change _dl_fini ordering at all).
The string parsing routine may end up writing beyond bounds of tunestr
if the input tunable string is malformed, of the form name=name=val.
This gets processed twice, first as name=name=val and next as name=val,
resulting in tunestr being name=name=val:name=val, thus overflowing
tunestr.
Terminate the parsing loop at the first instance itself so that tunestr
does not overflow.
This also fixes up tst-env-setuid-tunables to actually handle failures
correct and add new tests to validate the fix for this CVE.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Compilation fails when building with -DNDEBUG after commit a3189f66a5.
Here is the error:
dl-close.c: In function ‘_dl_close_worker’:
dl-close.c:140:22: error: unused variable ‘nloaded’ [-Werror=unused-variable]
140 | const unsigned int nloaded = ns->_ns_nloaded;
Add __attribute_maybe_unused__ for‘nloaded’to fix it.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Now binutils use some E_MIPS_* macros and EF_MIPS_* macros, it is
difficult to decide which style macro we should use when we want
to add new ELF file header flags.
IRIX used to use EF_MIPS_* macros and in elf/elf.h there also has
comments "The following are unofficial names and should not be used".
So we should use EF_MIPS_* to keep same style with the beginning.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
It is a left-over from commit 52a01100ad
("elf: Remove ad-hoc restrictions on dlopen callers [BZ #22787]").
When backporting commmit 6985865bc3
("elf: Always call destructors in reverse constructor order
(bug 30785)"), we can move the l_init_called_next field to this
place, so that the internal GLIBC_PRIVATE ABI does not change.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
The current implementation of dlclose (and process exit) re-sorts the
link maps before calling ELF destructors. Destructor order is not the
reverse of the constructor order as a result: The second sort takes
relocation dependencies into account, and other differences can result
from ambiguous inputs, such as cycles. (The force_first handling in
_dl_sort_maps is not effective for dlclose.) After the changes in
this commit, there is still a required difference due to
dlopen/dlclose ordering by the application, but the previous
discrepancies went beyond that.
A new global (namespace-spanning) list of link maps,
_dl_init_called_list, is updated right before ELF constructors are
called from _dl_init.
In dl_close_worker, the maps variable, an on-stack variable length
array, is eliminated. (VLAs are problematic, and dlclose should not
call malloc because it cannot readily deal with malloc failure.)
Marking still-used objects uses the namespace list directly, with
next and next_idx replacing the done_index variable.
After marking, _dl_init_called_list is used to call the destructors
of now-unused maps in reverse destructor order. These destructors
can call dlopen. Previously, new objects do not have l_map_used set.
This had to change: There is no copy of the link map list anymore,
so processing would cover newly opened (and unmarked) mappings,
unloading them. Now, _dl_init (indirectly) sets l_map_used, too.
(dlclose is handled by the existing reentrancy guard.)
After _dl_init_called_list traversal, two more loops follow. The
processing order changes to the original link map order in the
namespace. Previously, dependency order was used. The difference
should not matter because relocation dependencies could already
reorder link maps in the old code.
The changes to _dl_fini remove the sorting step and replace it with
a traversal of _dl_init_called_list. The l_direct_opencount
decrement outside the loader lock is removed because it appears
incorrect: the counter manipulation could race with other dynamic
loader operations.
tst-audit23 needs adjustments to the changes in LA_ACT_DELETE
notifications. The new approach for checking la_activity should
make it clearer that la_activty calls come in pairs around namespace
updates.
The dependency sorting test cases need updates because the destructor
order is always the opposite order of constructor order, even with
relocation dependencies or cycles present.
There is a future cleanup opportunity to remove the now-constant
force_first and for_fini arguments from the _dl_sort_maps function.
Fixes commit 1df71d32fe ("elf: Implement
force_first handling in _dl_sort_maps_dfs (bug 28937)").
Reviewed-by: DJ Delorie <dj@redhat.com>
In short: __tls_get_addr checks the global generation counter and if
the current dtv is older then _dl_update_slotinfo updates dtv up to the
generation of the accessed module. So if the global generation is newer
than generation of the module then __tls_get_addr keeps hitting the
slow dtv update path. The dtv update path includes a number of checks
to see if any update is needed and this already causes measurable tls
access slow down after dlopen.
It may be possible to detect up-to-date dtv faster. But if there are
many modules loaded (> TLS_SLOTINFO_SURPLUS) then this requires at
least walking the slotinfo list.
This patch tries to update the dtv to the global generation instead, so
after a dlopen the tls access slow path is only hit once. The modules
with larger generation than the accessed one were not necessarily
synchronized before, so additional synchronization is needed.
This patch uses acquire/release synchronization when accessing the
generation counter.
Note: in the x86_64 version of dl-tls.c the generation is only loaded
once, since relaxed mo is not faster than acquire mo load.
I have not benchmarked this. Tested by Adhemerval Zanella on aarch64,
powerpc, sparc, x86 who reported that it fixes the performance issue
of bug 19924.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Parts of elf/tst-rtld-list-diagnostics.py have been copied from
scripts/tst-ld-trace.py.
The abnf module is entirely optional and used to verify the
ABNF grammar as included in the manual.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Leads to build failures (preprocessor redefinitions), and there is not
enough time to address this properly. Deferred until after 2.38 release.
This reverts commit 59dc07637f.
Signed-off-by: Andreas K. Hüttel <dilfridge@gentoo.org>
Add new definitions for the MIPS target, specifically: relocation
types, machine flags, section type names, and object attribute tags
and values. On MIPS64, up to three relocations may be specified
within r_info, by the r_type, r_type2, and r_type3 fields, so add new
macros to get the respective reloc types for MIPS64.
Starting with commit 1bcfe0f732, the
test was enhanced and the object for __builtin_return_address (0)
is searched with _dl_find_object.
Unfortunately on e.g. s390 (31bit), a postprocessing step is needed
as the highest bit has to be masked out. This can be done with
__builtin_extract_return_addr.
Without this postprocessing, _dl_find_object returns with -1 and the
content of dlfo is invalid, which may lead to segfaults in basename.
Therefore those checks are now only done on success.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
The sparc ABI has multiple cases on how to handle JMP_SLOT relocations,
(sparc_fixup_plt/sparc64_fixup_plt). For BINDNOW, _dl_audit_symbind
will be responsible to setup the final relocation value; while for
lazy binding _dl_fixup/_dl_profile_fixup will call the audit callback
and tail cail elf_machine_fixup_plt (which will call
sparc64_fixup_plt).
This patch fixes by issuing the SPARC specific routine on bindnow and
forwarding the audit value to elf_machine_fixup_plt for lazy resolution.
It fixes the la_symbind for bind-now tests on sparc64 and sparcv9:
elf/tst-audit24a
elf/tst-audit24b
elf/tst-audit24c
elf/tst-audit24d
Checked on sparc64-linux-gnu and sparcv9-linux-gnu.
Tested-by: John Paul Adrian Glaubitz <glaubitz@physik.fu-berlin.de>
Success is reported with a 0 return value, and failure is -1.
Enhance the kitchen sink test elf/tst-audit28 to cover
_dl_find_object as well.
Fixes commit 5d28a8962d ("elf: Add _dl_find_object function")
and bug 30515.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Tested-by: Carlos O'Donell <carlos@redhat.com>
Add --enable-fortify-source option.
It is now possible to enable fortification through a configure option.
The level may be given as parameter, if none is provided, the configure
script will determine what is the highest level possible that can be set
considering GCC built-ins availability and set it.
If level is explicitly set to 3, configure checks if the compiler
supports the built-in function necessary for it or raise an error if it
isn't.
If the configure option isn't explicitly enabled, it _FORTIFY_SOURCE is
forcibly undefined (and therefore disabled).
The result of the configure checks are new variables, ${fortify_source}
and ${no_fortify_source} that can be used to appropriately populate
CFLAGS.
A dedicated patch will follow to make use of this variable in Makefiles
when necessary.
Updated NEWS and INSTALL.
Adding dedicated x86_64 variant that enables the configuration.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
The first segment in a shared library may be read-only, not executable.
To support LD_PREFER_MAP_32BIT_EXEC on such shared libraries, we also
check MAP_DENYWRITE to decide if MAP_32BIT should be passed to mmap.
Normally the first segment is mapped with MAP_COPY, which is defined
as (MAP_PRIVATE | MAP_DENYWRITE). But if the segment alignment is
greater than the page size, MAP_COPY isn't used to allocate enough
space to ensure that the segment can be properly aligned. Map the
first segment with MAP_COPY in this case to fix BZ #30452.
ldconfig was allocating PATH_MAX bytes on the stack for the library file
name. The issues with PATH_MAX usage are well documented [0][1]; even if
a program does not rely on paths being limited to PATH_MAX bytes,
allocating 4096 bytes on the stack for paths that are typically rather
short (strlen ("/lib64/libc.so.6") is 16) is wasteful and dangerous.
[0]: https://insanecoding.blogspot.com/2007/11/pathmax-simply-isnt.html
[1]: https://eklitzke.org/path-max-is-tricky
Instead, make use of asprintf to dynamically allocate memory of just the
right size on the heap.
Reviewed-by: Adhemerval Zanella <adhemerval.zanella@linaro.org>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Signed-off-by: Sergey Bugaev <bugaevc@gmail.com>
With fortification enabled, system calls return result needs to be checked,
has it gets the __wur macro enabled.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Sort Makefile variables using scrips/sort-makefile-lines.py.
No code generation changes observed in non-test binary artifacts.
No regressions on x86_64 and i686.
Reviewed-by: Florian Weimer <fweimer@redhat.com>
Previously, after destructors for a DSO have been invoked, ld.so refused
to bind against that DSO in all cases. Relax this restriction somewhat
if the referencing object is itself a DSO that is being unloaded. This
assumes that the symbol reference is not going to be stored anywhere.
The situation in the test case can arise fairly easily with C++ and
objects that are built with different optimization levels and therefore
define different functions with vague linkage.
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Applying this commit results in bit-identical libc.so.6.
The elf/ld-linux-x86-64.so.2 does change, but only in .note.gnu.build-id
Reviewed-by: Carlos O'Donell <carlos@redhat.com>
Linux 6.3 adds constants AT_RSEQ_FEATURE_SIZE and AT_RSEQ_ALIGN; add
them to glibc's elf.h. (Recall that, although elf.h is a
system-independent header, so far we've put AT_* constants there even
if Linux-specific, as discussed in bug 15794. So rather than making
any attempt to fix that issue, the new constants are just added there
alongside the existing ones.)
Tested for x86_64.
This patch checks _dl_debug_vdprintf, by passing various inputs to
_dl_dprintf and comparing the output with invocations of snprintf.
Signed-off-by: Roy Eldar <royeldar0@gmail.com>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
_dl_debug_vdprintf is a bare-bones printf implementation; currently
printing a signed integer (using "%d" format specifier) behaves
incorrectly when the number is negative, as it just prints the
corresponding unsigned integer, preceeded by a minus sign.
For example, _dl_printf("%d", -1) would print '-4294967295'.
Signed-off-by: Roy Eldar <royeldar0@gmail.com>
Reviewed-by: Florian Weimer <fweimer@redhat.com>
f55727ca53 updated open_path to use the
r_search_path_struct struct but failed to update the comment.
Signed-off-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
When dlopen is being called, efforts have been made to improve
future lookup performance. This includes marking a search path
as non-existent using `stat`. However, if the root directory
is given as a search path, there exists a bug which erroneously
marks it as non-existing.
The bug is reproduced under the following sequence:
1. dlopen is called to open a shared library, with at least:
1) a dependency 'A.so' not directly under the '/' directory
(e.g. /lib/A.so), and
2) another dependency 'B.so' resides in '/'.
2. for this bug to reproduce, 'A.so' should be searched *before* 'B.so'.
3. it first tries to find 'A.so' in /, (e.g. /A.so):
- this will (obviously) fail,
- since it's the first time we have seen the '/' directory,
its 'status' is 'unknown'.
4. `buf[buflen - namelen - 1] = '\0'` is executed:
- it intends to remove the leaf and its final slash,
- because of the speciality of '/', its buflen == namelen + 1,
- it erroneously clears the entire buffer.
6. it then calls 'stat' with the empty buffer:
- which will result in an error.
7. so it marks '/' as 'nonexisting', future lookups will not consider
this path.
8. while /B.so *does* exist, failure to look it up in the '/'
directory leads to a 'cannot open shared object file' error.
This patch fixes the bug by preventing 'buflen', an index to put '\0',
from being set to 0, so that the root '/' is always kept.
Relative search paths are always considered as 'existing' so this
wont be affected.
Writeup by Moody Liu <mooodyhunter@outlook.com>
Suggested-by: Carlos O'Donell <carlos@redhat.com>
Signed-off-by: Qixing ksyx Xue <qixingxue@outlook.com>
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Sort tests against updated scripts/sort-makefile-lines.py.
No changes in generated code.
No regressions on x86_64 and i686.
Reviewed-by: Siddhesh Poyarekar <siddhesh@sourceware.org>
Fix list terminator whitspace.
Sort using scripts/sort-makefile-lines.py.
No code generation changes observed in binary artifacts.
No regressions on x86_64 and i686.
