om> wrote:
>
>
> On Wed, Jan 31, 2018 at 1:33 PM, Francis Ricci <francisjri...@gmail.com>
> wrote:
>
>> * Most system allocators have inline metadata (i.e. 8 or 16 bytes to
>> the left of the chunk returned to the user). Those will have to be
>> poisoned somehow. Also,
e heap chunks (aka
> madvise don't need)?
> * How do you report bugs with detailed position inside the heap chunk?
> * probably a couple more
>
>
> With our new ASAN (HWASAN,
> clang.llvm.org/docs/HardwareAssistedAddressSanitizerDesign.html)
> most of these things will become much simpler since there
Hi all,
I may be missing something conceptually with the way ASan works, but
is there any reason that ASan couldn't pass-through allocations to the
user's allocator? For example:
1) User calls malloc()
2) ASan intercepts malloc(), does checks/adds metadata, etc
3) Instead of using the sanitizer
interesting to see the effect of not instrumenting globals (-mllvm
> -asan-globals=0)
> and not instrumenting stack (-mllvm -asan-stack=0), separately.
>
> Thanks!
>
> --kcc
>
>
> On Fri, Sep 29, 2017 at 10:56 AM, Francis Ricci <francisjri...@gmail.com>
> wrote:
>&
Ahh nvm, I misunderstood what the in-process symbolication was. Didn't
realize that you need to build llvm-symbolizer either way. Thanks!
On Thu, Aug 24, 2017, 3:26 PM Francis Ricci <francisjri...@gmail.com> wrote:
> It would be less painful primarily because my codebase already ha
Hi all,
Is there currently a good way to run ASan/LSan on a program using a custom
allocator (for example, tcmalloc)? I couldn't find any solutions in the
current code other than manually adding some extra interceptors locally (in
the tcmalloc case those would be things like tc_malloc and
Thanks!
On Thu, May 25, 2017 at 5:24 PM, 'Evgenii Stepanov' via
address-sanitizer <address-sanitizer@googlegroups.com> wrote:
> Yes, you need this fix:
> https://android-review.googlesource.com/#/c/318859/
>
>
> On Wed, May 24, 2017 at 11:41 AM, Francis Ricci <francisjr
Appears to work fine on android 6, so my guess is that 7 introduces a
bug of some sort.
On Wed, May 24, 2017 at 11:57 AM, Francis Ricci <francisjri...@gmail.com> wrote:
> Do you have experience using the wrap property? When I attempt to use it,
> even for very simple wrappers on
evice based on the app manifest - some property like
> "debuggable". That requires changing android platform source.
>
> https://source.android.com/devices/tech/debug/asan#using_the_wrap_property
>
> On Thu, May 18, 2017 at 8:29 AM, Francis Ricci <franci...@gmail
platform source.
That seems like it would be even better, although I question whether
android would be receptive to that sort of change,
given potential security implications, etc.
>>
>> https://source.android.com/devices/tech/debug/asan#using_the_wrap_property
>>
>> On T
Hi all!
Historically, I've always used ASan on a rooted/flashed device, as the
asan_device_setup script requires root access. Is it possible (even in
theory, or with some extra work in the codebase) to run ASan on a
non-rooted device? I know you can for a simple C++ binary, since you can
just
Hi all,
I'm currently working on porting LSAN to Darwin 64-bit targets, and I've
begun uploading patches to support that. Currently, support is very much in
progress, and I'm only able to run successfully on small, simple test
cases. However, I'm committed to developing and adding support for
Hi all,
I'm currently working on porting LSAN to Darwin 64-bit targets, and I've
begun uploading patches to support that. Currently, support is very much in
progress, and I'm only able to run successfully on small, simple test
cases. However, I'm committed to developing and adding support for
13 matches
Mail list logo
