by explicitly listing where we expect this test to work.
Because this test invokes undefined behaviour all sorts of things
can happen (e.g. crash, or on some bots asan manages to catch
something). The test seems to pass okay on Darwin (x86_64/x86_64/i386)
and on Linux (x86_64/i386) so explicitly require one of these platforms
to run the test.
llvm-svn: 327185
This fixes a false positive ODR violation that is reported by ASan when using LTO. In cases, where two constant globals have the same value, LTO will merge them, which breaks ASan's ODR detection. See the included testcase for an example.
Differential Revision: https://reviews.llvm.org/D43959
llvm-svn: 327062
This fixes a false positive ODR violation that is reported by ASan when using LTO. In cases, where two constant globals have the same value, LTO will merge them, which breaks ASan's ODR detection.
Differential Revision: https://reviews.llvm.org/D43959
llvm-svn: 327031
Summary:
Fixes Bug 32434
See https://bugs.llvm.org/show_bug.cgi?id=32434
Short summary:
std::rethrow_exception does not use __cxa_throw to rethrow the exception, so if
it is called from uninstrumented code, it will leave the stack poisoned. This
can lead to false positives.
Long description:
For functions which don't return normally (e.g. via exceptions), asan needs to
unpoison the entire stack. It is not known before a call to such a function
where execution will continue, some function which don't contain cleanup code
like destructors might be skipped. After stack unwinding, execution might
continue in uninstrumented code.
If the stack has been poisoned before such a function is called, but the stack
is unwound during the unconventional return, then zombie redzones (entries) for
no longer existing stack variables can remain in the shadow memory. Normally,
this is avoided by asan generating a call to asan_handle_no_return before all
functions marked as [[noreturn]]. This asan_handle_no_return unpoisons the
entire stack. Since these [[noreturn]] functions can be called from
uninstrumented code, asan also introduces interceptor functions which call
asan_handle_no_return before running the original [[noreturn]] function;
for example, cxa_throw is intercepted.
If a [[noreturn]] function is called from uninstrumented code (so the stack is
left poisoned) and additionally, execution continues in uninstrumented code, new
stack variables might be introduced and overlap with the stack variables
which have been removed during stack unwinding. Since the redzones are not
cleared nor overwritten by uninstrumented code, they remain but now contain
invalid data.
Now, if the redzones are checked against the new stack variables, false
positive reports can occur. This can happen for example by the uninstrumented
code calling an intercepted function such as memcpy, or an instrumented
function.
Intercepting std::rethrow_exception directly is not easily possible since it
depends on the C++ standard library implementation (e.g. libcxx vs libstdc++)
and the mangled name it produces for this function. As a rather simple
workaround, we're intercepting _Unwind_RaiseException for libstdc++. For
libcxxabi, we can intercept the ABI function __cxa_rethrow_primary_exception.
Patch by Robert Schneider.
Reviewers: kcc, eugenis, alekseyshl, vitalybuka
Reviewed By: vitalybuka
Subscribers: llvm-commits
Differential Revision: https://reviews.llvm.org/D42644
llvm-svn: 326132
It looks like get_bits_for_arch doesn't recognize "arm64" as a 64-bit architecture, and it actually matches the "arm" regexp, which marks it as 32-bit. Let's fix that by matching the 64-bit list first and addin "arm64" into the list.
Differential Revision: https://reviews.llvm.org/D43155
llvm-svn: 325300
The "sleep(5)" sometimes times out on our bots, causing the test to fail. Let's use pthread_join.
Differential Revision: https://reviews.llvm.org/D42862
llvm-svn: 324126
Summary:
With the change, one can choose not to report comparison (or subtraction)
of a pointer with nullptr pointer.
Reviewers: kcc, jakubjelinek, alekseyshl
Reviewed By: alekseyshl
Subscribers: kubamracek
Differential Revision: https://reviews.llvm.org/D41479
llvm-svn: 323995
In more recent Linux kernels with 47 bit VMAs the layout of virtual memory
for powerpc64 changed causing the address sanitizer to not work properly. This
patch fixes up a test case that was found to fail on some newer Fedora
releases that use different address ranges.
ref: https://reviews.llvm.org/D40907
llvm-svn: 323217
Summary:
One test-case uses a wrong operation (should be subtraction).
Second test-case should declare a global variables before a tested one
in order to guarantee we will find a red-zone.
Reviewers: kcc, jakubjelinek, alekseyshl
Reviewed By: alekseyshl
Subscribers: kubamracek
Differential Revision: https://reviews.llvm.org/D41481
llvm-svn: 323162
sequentially.
The current implementation of commands in
`test/sanitizer_common/ios_commands/` for iOS devices cannot be executed
in parallel which results in the ASan and TSan tests failing when
executed in parallel by lit which was the default behaviour.
We now force the ASan and TSan tests to be a new parallelism group named
`darwin-ios-device-sanitizer` which allows only one test to be run at a
time. We also emit a warning informing the user that tests are being
run sequentially.
This only applies if the target is an iOS device.
Differential Revision: https://reviews.llvm.org/D42156
llvm-svn: 323026
Summary:
This patch (on top of the previous two (https://reviews.llvm.org/D40898 and
https://reviews.llvm.org/D40899) complete the compiler-rt side of the the Solaris
sanitizer port.
It contains the following sets of changes:
* For the time being, the port is for 32-bit x86 only, so reject the various tests on
x86_64.
* When compiling as C++, <setjmp.h> resp. <iso/setjmp_iso.h> only declares
_setjmp and _longjmp inside namespace std.
* MAP_FILE is a Windows feature. While e.g. Linux <sys/mman.h> provides a
no-op compat define, Solaris does not.
* test/asan/TestCases/Posix/coverage.cc was initially failing like this:
/vol/gcc/src/llvm/llvm/local/projects/compiler-rt/lib/sanitizer_common/scripts/sancov.py: 4 files merged; 2 PCs total
rm: cannot remove '/var/gcc/llvm/local/projects/compiler-rt/test/asan/I386SunOSConfig/TestCases/Posix/Output/coverage': Invalid argument
Further digging revealed that the rm was trying to remove the running test's working
directory which failed as observed. cd'ing out of the dir before let the test pass.
* Two tests needed a declaration of alloca. I've now copied the existing code from
test/asan/TestCases/alloca_constant_size.cc, but it may be more profitable and
maintainable to have a common testsuite header where such code is collected.
* Similarly, Solaris' printf %p format doesn't include the leading 0x.
* In test/asan/TestCases/malloc-no-intercept.c, I had to undef __EXTENSIONS__
(predefined by clang for no apparent reason) to avoid conflicting declarations
for memalign.
* test/ubsan/TestCases/Float/cast-overflow.cpp has different platform dependent
ways to define BYTE_ORDER and friends. Why not just use __BYTE_ORDER__ and
friends as predefined by clang and gcc?
Patch by Rainer Orth.
Reviewers: kcc, alekseyshl
Reviewed By: alekseyshl
Subscribers: srhines, kubamracek, mgorny, krytarowski, fedor.sergeev, JDevlieghere, llvm-commits, #sanitizers
Tags: #sanitizers
Differential Revision: https://reviews.llvm.org/D40900
llvm-svn: 322635
Summary:
Export aligned new/delete to make dynamic runtimes work again.
Remove all valid new/delete cases from ASan test, there's a test in
common for that.
Reviewers: eugenis
Subscribers: srhines, kubamracek, #sanitizers, llvm-commits
Differential Revision: https://reviews.llvm.org/D41548
llvm-svn: 321394
There could be a situation when a specific DSO was built with FORTIFY_SOURCE option. In case asan-ed binary link against that DSO,
libasan can't handle the possible memory error because it does not have interceptors for spinrtf_chk, snprintf_chk, vprintf_chk,
vsnprintf_chk, __fprintf_chk functions. Let's interceptors for them.
Patch by Denis Khalikov.
Differential Revision: https://reviews.llvm.org/D40951
llvm-svn: 320990
In more recent Linux kernels with 47 bit VMAs the layout of virtual memory
for powerpc64 changed causing the address sanitizer to not work properly. This
patch adds support for 47 bit VMA kernels for powerpc64 and fixes up test
cases.
https://reviews.llvm.org/D40908
There is an associated patch for trunk.
Tested on several 4.x and 3.x kernel releases.
llvm-svn: 320110
Following patch adds support of all memory origins in
CheckForInvalidPointerPair function. For small difference of pointers,
it's directly done in shadow memory (the limit was set to 2048B).
Then we search for origin of first pointer and verify that the second
one has the same origin. If so, we verify that it points either to a same
variable (in case of stack memory or a global variable), or to a same
heap segment.
Committing on behanf of marxin and jakubjelinek.
Reviewers: alekseyshl, kcc
Subscribers: llvm-commits
Differential revision: https://reviews.llvm.org/D40600
llvm-svn: 319668
Summary:
As discussed in https://github.com/google/oss-fuzz/issues/933,
it would be really awesome to be able to use ThinLTO for fuzzing.
However, as @kcc has pointed out, it is currently undefined (untested)
whether the sanitizers actually function properly with LLD and/or LTO.
This patch is inspired by the cfi test, which already do test with LTO
(and/or LLD), since LTO is required for CFI to function.
I started with UBSan, because it's cmakelists / lit.* files appeared
to be the cleanest. This patch adds the infrastructure to easily add
LLD and/or LTO sub-variants of the existing lit test configurations.
Also, this patch adds the LLD flavor, that explicitly does use LLD to link.
The check-ubsan does pass on my machine. And to minimize the [initial]
potential buildbot breakage i have put some restrictions on this flavour.
Please review carefully, i have not worked with lit/sanitizer tests before.
The original attempt, r319525 was reverted in r319526 due
to the failures in compiler-rt standalone builds.
Reviewers: eugenis, vitalybuka
Reviewed By: eugenis
Subscribers: #sanitizers, pcc, kubamracek, mgorny, llvm-commits, mehdi_amini, inglorion, kcc
Differential Revision: https://reviews.llvm.org/D39508
llvm-svn: 319575
This reverts commit r319525.
This change has introduced a problem with the Lit tests build for compiler-rt using Gold: http://lab.llvm.org:8011/builders/sanitizer-x86_64-linux/builds/6047/steps/test%20standalone%20compiler-rt/logs/stdio
llvm-lit: /b/sanitizer-x86_64-linux/build/llvm/utils/lit/lit/TestingConfig.py:101: fatal: unable to parse config file '/b/sanitizer-x86_64-linux/build/llvm/projects/compiler-rt/test/profile/Linux/lit.local.cfg', traceback: Traceback (most recent call last):
File "/b/sanitizer-x86_64-linux/build/llvm/utils/lit/lit/TestingConfig.py", line 88, in load_from_path
exec(compile(data, path, 'exec'), cfg_globals, None)
File "/b/sanitizer-x86_64-linux/build/llvm/projects/compiler-rt/test/profile/Linux/lit.local.cfg", line 37, in <module>
if root.host_os not in ['Linux'] or not is_gold_linker_available():
File "/b/sanitizer-x86_64-linux/build/llvm/projects/compiler-rt/test/profile/Linux/lit.local.cfg", line 27, in is_gold_linker_available
stderr = subprocess.PIPE)
File "/usr/lib/python2.7/subprocess.py", line 390, in __init__
errread, errwrite)
File "/usr/lib/python2.7/subprocess.py", line 1024, in _execute_child
raise child_exception
OSError: [Errno 2] No such file or directory
llvm-svn: 319529
Summary:
As discussed in https://github.com/google/oss-fuzz/issues/933,
it would be really awesome to be able to use ThinLTO for fuzzing.
However, as @kcc has pointed out, it is currently undefined (untested)
whether the sanitizers actually function properly with LLD and/or LTO.
This patch is inspired by the cfi test, which already do test with LTO
(and/or LLD), since LTO is required for CFI to function.
I started with UBSan, because it's cmakelists / lit.* files appeared
to be the cleanest. This patch adds the infrastructure to easily add
LLD and/or LTO sub-variants of the existing lit test configurations.
Also, this patch adds the LLD flavor, that explicitly does use LLD to link.
The check-ubsan does pass on my machine. And to minimize the [initial]
potential buildbot breakage i have put some restrictions on this flavour.
Please review carefully, i have not worked with lit/sanitizer tests before.
Reviewers: eugenis, vitalybuka
Reviewed By: eugenis
Subscribers: #sanitizers, pcc, kubamracek, mgorny, llvm-commits, mehdi_amini, inglorion, kcc
Differential Revision: https://reviews.llvm.org/D39508
llvm-svn: 319525
On macOS, we usually don't require launching the target with DYLD_INSERT_LIBRARIES anymore. However, it is still necessary when running a target that is not instrumented (and e.g. dlopen's an instrument library later). In any case, ASan and TSan currently remove themselves from the DYLD_INSERT_LIBRARIES environment variable to avoid passing it onto children. This works well e.g. when instrumenting a shell. A problem arises when the target is a non-instrumented shim (e.g. "xcrun") that either re-execs or launches a child that is supposed to get DYLD_INSERT_LIBRARIES propagated. To support this mode, this patch introduces 'strip_env' flag that can be used to keep DYLD_INSERT_LIBRARIES untouched.
Differential Revision: https://reviews.llvm.org/D39991
llvm-svn: 319365
It's explicitly forbidden to call fclose with NULL, but at least on Darwin, this succeeds and doesn't segfault. To maintain binary compatibility, ASan should survice fclose(NULL) as well.
Differential Revision: https://reviews.llvm.org/D40053
llvm-svn: 319347
Calling getpwnam(NULL) is probably a bug, but at least on Darwin, such a call succeeds without segfaulting. I have some existing code that relies on that. To maintain binary compatibility, ASan should also survive a call to getpwnam with NULL.
Differential Revision: https://reviews.llvm.org/D40052
llvm-svn: 319344
PDB emission now works well enough that we can rely on it for these
tests to pass.
Differential Revision: https://reviews.llvm.org/D40188
llvm-svn: 318546
The tests are ported as follows:
contiguous_container_crash.cc
use-after-delete.cc
use-after-free.cc
Replace hardwired shadow granularity in CHECK statements with regex.
max_redzone.cc
Bump max_redzone parameter to 32.
memset_test.cc
Bump size parameter of __asan_poison_memory_region to 32.
scariness_score_test.cc
For "far-from-bounds" heap overflow, make sure overflow is more than
one shadow granularity away.
At large shadow granularity, there is not enough redzone between
stack elements to detect far-from-bounds, so fake out that test.
Differential Revision: https://reviews.llvm.org/D39773
llvm-svn: 318470
ASan allocator stores the requested alignment for new and new[] calls
and on delete and delete[] verifies that alignments do match.
The representable alignments are: default alignment, 8, 16, 32, 64, 128,
256 and 512 bytes. Alignments > 512 are stored as 512, hence two
different alignments > 512 will pass the check (possibly masking the bug),
but limited memory requirements deemed to be a resonable tradeoff for
relaxed conditions.
The feature is controlled by new_delete_type_mismatch flag, the same one
protecting new/delete matching size check.
Differential revision: https://reviews.llvm.org/D38574
Issue: https://github.com/google/sanitizers/issues/799
llvm-svn: 316595
Summary:
Purging allocator quarantine and returning memory to OS might be desired
between fuzzer iterations since, most likely, the quarantine is not
going to catch bugs in the code under fuzz, but reducing RSS might
significantly prolong the fuzzing session.
Reviewers: cryptoad
Subscribers: kubamracek, llvm-commits
Differential Revision: https://reviews.llvm.org/D39153
llvm-svn: 316347
This is a very poorly named feature. I think originally it meant to cover linux only, but the use of it in msan
seems to be about any aarch64 platform. Anyway, this change should be NFC on everything except Android.
llvm-svn: 315389
compunit's .data section. This vector is not poisoned. Because of this the
first symbol of the following section has no left red zone. As a result, ASan
cannot detect underflow for such symbols.
Poison ASan allocated metadata, it should not be accessible to user code.
This fix does not eliminate the problem with missing left red zones but it
reduces the set of vulnerable symbols from first symbols in each input data
section to first symbols in the output section of the binary.
Differential Revision: https://reviews.llvm.org/D38056
llvm-svn: 314365
Linux for mips has a non-standard layout for the kernel sigaction struct.
Adjust the layout by the minimally amount to get the test to pass, as we
don't require the usage of the restorer function.
llvm-svn: 314200
Don't overwrite exit code in LSan when running on top of ASan in recovery mode
to avoid breakage of users code due to found leaks.
Patch by Slava Barinov.
Differential Revision: https://reviews.llvm.org/D38026
llvm-svn: 313966
Summary:
1. Update ubsan_interface.inc to make the test happy.
2. Switch interface_symbols_linux and interface_symbols_darwin to C++ to import __ubsan_handle_dynamic_type_cache_miss
3. Switch interface_symbols_windows to C++ for consistency.
Reviewers: rnk, zturner
Subscribers: llvm-commits, kubamracek
Differential Revision: https://reviews.llvm.org/D37986
llvm-svn: 313551
Remove the explicit i686 target that is completely duplicate to
the i386 target, with the latter being used more commonly.
1. The runtime built for i686 will be identical to the one built for
i386.
2. Supporting both -i386 and -i686 suffixes causes unnecessary confusion
on the clang end which has to expect either of them.
3. The checks are based on wrong assumption that __i686__ is defined for
all newer x86 CPUs. In fact, it is only declared when -march=i686 is
explicitly used. It is not available when a more specific (or newer)
-march is used.
Curious enough, if CFLAGS contain -march=i686, the runtime will be built
both for i386 and i686. For any other value, only i386 variant will be
built.
Differential Revision: https://reviews.llvm.org/D26764
llvm-svn: 311924
Remove the explicit i686 target that is completely duplicate to
the i386 target, with the latter being used more commonly.
1. The runtime built for i686 will be identical to the one built for
i386.
2. Supporting both -i386 and -i686 suffixes causes unnecessary confusion
on the clang end which has to expect either of them.
3. The checks are based on wrong assumption that __i686__ is defined for
all newer x86 CPUs. In fact, it is only declared when -march=i686 is
explicitly used. It is not available when a more specific (or newer)
-march is used.
Curious enough, if CFLAGS contain -march=i686, the runtime will be built
both for i386 and i686. For any other value, only i386 variant will be
built.
Differential Revision: https://reviews.llvm.org/D26764
llvm-svn: 311842
Summary:
`pvalloc` appears to not be available on Android. Mark the failing test as
unsupported on that platform.
Reviewers: alekseyshl, vitalybuka
Reviewed By: alekseyshl, vitalybuka
Subscribers: srhines, kubamracek, llvm-commits
Differential Revision: https://reviews.llvm.org/D36339
llvm-svn: 310133
Summary:
Last one of the `pvalloc` overflow checks!
`CheckForPvallocOverflow` was introduced with D35818 to detect when `pvalloc`
would wrap when rounding up to the next multiple of the page size.
Add this check to ASan's `pvalloc` implementation.
Reviewers: alekseyshl
Reviewed By: alekseyshl
Subscribers: llvm-commits, kubamracek
Differential Revision: https://reviews.llvm.org/D36257
llvm-svn: 310119
Summary:
Included is one test for passing structs by value and one test for
passing C++
objects by value.
Reviewers: eugenis, vitalybuka
Reviewed By: eugenis
Subscribers: srhines, kubamracek, llvm-commits
Differential Revision: https://reviews.llvm.org/D34827
llvm-svn: 309424
atos is the default symbolizer on Apple's compiler for quite a few years now.
llvm-symbolizer is quite fragile on Darwin: for example, unless a .dSYM
file was explicitly generated symbolication would not work.
It is also very convenient when the behavior of LLVM open source
compiler matches to that of Apple's compiler on Apple's platform.
Furthermore, llvm-symbolizer is not installed on Apple's platform by
default, which leads to strange behavior during debugging: the test
might fail under lit (where it has llvm-symbolizer) but would run
properly when launched on the command line (where it does not, and atos
would be used).
Indeed, there's a downside: atos does not work properly with inlined
functions, hence the test change.
We do not think that this is a major problem, as users would often
compile with -O0 when debugging, and in any case it is preferable to
symbolizer not being able to symbolize.
Differential Revision: https://reviews.llvm.org/D35745
llvm-svn: 308908
Summary:
Set proper errno code on alloction failures and change some
implementations to satisfy their man-specified requirements:
LSan: valloc and memalign
ASan: pvalloc, memalign and posix_memalign
Changing both allocators in one patch since LSan depends on ASan allocator in some configurations.
Reviewers: vitalybuka
Subscribers: kubamracek, llvm-commits
Differential Revision: https://reviews.llvm.org/D35440
llvm-svn: 308064
Revert "Copy arguments passed by value into explicit allocas for ASan."
Revert "[asan] Add end-to-end tests for overflows of byval arguments."
Build failure on lldb-x86_64-ubuntu-14.04-buildserver.
Test failure on clang-cmake-aarch64-42vma and sanitizer-x86_64-linux-android.
llvm-svn: 307345
Included is one test for passing structs by value and one test for passing C++
objects by value.
Patch by Matt Morehouse.
Differential revision: https://reviews.llvm.org/D34827
llvm-svn: 307343
Evgeniy Stepanov