Commit Graph

55 Commits

Author SHA1 Message Date
Kostya Serebryany 43c4493b44 [asan] second attempt to use TLS with fake stack. This time it looks (more) async-signal safe.
llvm-svn: 190663
2013-09-13 06:32:26 +00:00
Kostya Serebryany 44441cc34b [asan] (part 2) don't lazy-init fake_stack if we only need to check that fake_stack exists (should fix 32-bit builds)
llvm-svn: 190594
2013-09-12 08:47:00 +00:00
Kostya Serebryany 736bd08e05 [asan] don't lazy-init fake_stack if we only need to check that fake_stack exists (should fix 32-bit builds)
llvm-svn: 190593
2013-09-12 08:43:44 +00:00
Kostya Serebryany 628cda7367 [asan] hopefully make the FakeStack async-signal safe, enable the related test
llvm-svn: 190592
2013-09-12 08:34:50 +00:00
Alexey Samsonov 4b16885839 Improve collecting malloc stats in ASan
Summary:
This change makes races between updates of thread-local stats and
merging all the thread-local stats together less harmful.

Reviewers: kcc

Reviewed By: kcc

CC: dvyukov, llvm-commits

Differential Revision: http://llvm-reviews.chandlerc.com/D1572

llvm-svn: 189744
2013-09-02 08:39:07 +00:00
Sergey Matveev bdeff959a1 [lsan] Handle fork() correctly.
Update the main thread's os_id on every pthread_create, and before
initiating leak checking. This ensures that we have the correct os_id even if we
have forked after Init().

llvm-svn: 185815
2013-07-08 12:57:24 +00:00
Kostya Serebryany 6a068a715d [asan] initialize fake_stack lazily and increase its maximal size. This makes -fsanitize=address,use-after-return more robust: all SPEC tests pass now. In the default mode thread stacks become a bit smaller.
llvm-svn: 184934
2013-06-26 12:16:05 +00:00
Sergey Matveev b94d5e2d1c [asan] Move lsan_disabled out of thread context.
Fix for the case where disabler is used in pthread key destructor.

llvm-svn: 184553
2013-06-21 14:51:52 +00:00
Timur Iskhodzhanov baf90ccfac Fix ALIGNED misuse in asan_thread.cc (built on all platforms); also, add a comment to the ALIGNED macro describing the correct usage
llvm-svn: 183214
2013-06-04 08:25:17 +00:00
Kostya Serebryany c1aa0e8f69 [asan] ASan Linux MIPS32 support (compiler-rt part), patch by Jyun-Yan Y
llvm-svn: 183105
2013-06-03 14:49:25 +00:00
Sergey Matveev 17ee1abfa7 [lsan] Add __lsan_disable() and __lsan_enable().
Objects allocated after a call to __lsan_disable() will be treated as
live memory. Also add a ScopedDisabler.

llvm-svn: 183099
2013-06-03 11:21:34 +00:00
Sergey Matveev 09886cd17a [asan] Make ASan report the correct thread address ranges to LSan.
This CL enables thread support in LSan when used on top of ASan.

llvm-svn: 182854
2013-05-29 13:09:44 +00:00
Sergey Matveev 65dd62abe5 [asan] LSan hooks in asan_thread.cc
llvm-svn: 182378
2013-05-21 13:40:13 +00:00
Kostya Serebryany 5b4267f7e7 [sanitizer] found a bug by code inspection: CHECK(a=b) instead of CHECK(a==b). Was puzzled why lint did not catch it. Turns out this check was disabled for asan source. fix all cases and enable the check
llvm-svn: 178872
2013-04-05 14:40:25 +00:00
Alexey Samsonov a88c60b085 [ASan] Speed-up initialization-order checking: create and use fast versions of PoisonShadow functions, store copies of __asan_global descriptors in a vector instead of list of pointers. This gives 3x speedup on both benchmarks and real binaries with lots of globals.
llvm-svn: 178239
2013-03-28 15:42:43 +00:00
Kostya Serebryany 667a34a120 [asan] Change the way we report the alloca frame on stack-buff-overflow.
Before: the function name was stored by the compiler as a constant string
and the run-time was printing it.
Now: the PC is stored instead and the run-time prints the full symbolized frame.
This adds a couple of instructions into every function with non-empty stack frame,
but also reduces the binary size because we store less strings (I saw 2% size reduction).
This change bumps the asan ABI version to v3.

compiler-rt part, llvm part will follow.

Example of report (now):
==31711==ERROR: AddressSanitizer: stack-buffer-overflow on address 0x7fffa77cf1c5 at pc 0x41feb0 bp 0x7fffa77cefb0 sp 0x7fffa77cefa8
READ of size 1 at 0x7fffa77cf1c5 thread T0
    #0 0x41feaf in Frame0(int, char*, char*, char*) stack-oob-frames.cc:20
    #1 0x41f7ff in Frame1(int, char*, char*) stack-oob-frames.cc:24
    #2 0x41f477 in Frame2(int, char*) stack-oob-frames.cc:28
    #3 0x41f194 in Frame3(int) stack-oob-frames.cc:32
    #4 0x41eee0 in main stack-oob-frames.cc:38
    #5 0x7f0c5566f76c (/lib/x86_64-linux-gnu/libc.so.6+0x2176c)
    #6 0x41eb1c (/usr/local/google/kcc/llvm_cmake/a.out+0x41eb1c)
Address 0x7fffa77cf1c5 is located in stack of thread T0 at offset 293 in frame
    #0 0x41f87f in Frame0(int, char*, char*, char*) stack-oob-frames.cc:12  <<<<<<<<<<<<<< this is new
  This frame has 6 object(s):
    [32, 36) 'frame.addr'
    [96, 104) 'a.addr'
    [160, 168) 'b.addr'
    [224, 232) 'c.addr'
    [288, 292) 's'
    [352, 360) 'd'

llvm-svn: 177723
2013-03-22 10:36:24 +00:00
Dmitry Vyukov a7e42b5be5 asan: fix lint warning about line length
llvm-svn: 177709
2013-03-22 07:29:59 +00:00
Alexey Samsonov 54afba8b62 [ASan] Switch ASan to generic ThreadRegistry from sanitizer_common. Delete ASan-specific AsanThreadRegistry.
llvm-svn: 177634
2013-03-21 11:23:41 +00:00
Alexey Samsonov cf025cb2e6 [ASan] Move GetCurrentThread/SetCurrentThread from AsanThreadRegistry class into plain functions: they don't actually use registry
llvm-svn: 177501
2013-03-20 09:23:28 +00:00
Kostya Serebryany 63c36bbe5e [asan] fix two off-by-one errors that seem to affect only PowerPC because only there the stack top may be equal to the address space top. Noted by Andreas Schwab in http://gcc.gnu.org/bugzilla/show_bug.cgi?id=55975#c11 . Also make swapcontext interceptor a bit more robust
llvm-svn: 172807
2013-01-18 11:30:36 +00:00
Kostya Serebryany f22c697f58 [asan] get rid of some of the uses of kPageSize. The intent is to get rid of it completely to support platforms with multiple possible page sizes.
llvm-svn: 168517
2012-11-23 15:38:49 +00:00
Kostya Serebryany 734f1eb5f4 [asan/tsan] do not use __WORDSIZE macro, as it is glibc-private thing. Instead, define our own SANITIZER_WORDSIZE
llvm-svn: 168424
2012-11-21 12:38:58 +00:00
Alexander Potapenko bcc00a4174 [ASan] Revert r168040 and r168043 and take a cleaner solution suggested by Kostya: return the known frame name for fake stack instead of looking it up.
llvm-svn: 168046
2012-11-15 15:24:42 +00:00
Alexander Potapenko 93dcc8d2bd [ASan] Minor fix: return to the last byte of the fake stack if we've raced by.
llvm-svn: 168043
2012-11-15 14:59:19 +00:00
Alexander Potapenko b34db9e883 [ASan] Poison the leftmost shadow byte with a special value so that we can find
the beginning of the fake frame when reporting an use-after-return error.
Fixes http://code.google.com/p/address-sanitizer/issues/detail?id=126

llvm-svn: 168040
2012-11-15 13:40:44 +00:00
Kostya Serebryany f9caa28ccf [asan] increase max stack size to 256 (+test)
llvm-svn: 163308
2012-09-06 10:57:03 +00:00
Kostya Serebryany 6b0d775229 [asan] some renaming before we move StackTrace into sanitizer_common
llvm-svn: 162747
2012-08-28 11:54:30 +00:00
Alexey Samsonov e67a5dd8e1 [Sanitizer] Use low-level allocator in flag parsing to avoid calling malloc() before ASan/TSan initialization is done
llvm-svn: 162673
2012-08-27 14:04:54 +00:00
Alexander Potapenko 51e6488b31 Intercept CFAllocator for each thread in the program.
Test that child threads use the ASan allocator, that allocated memory can be passed to another thread and deallocated on it.
This should fix http://code.google.com/p/address-sanitizer/issues/detail?id=81

llvm-svn: 160630
2012-07-23 14:07:58 +00:00
Kostya Serebryany 78713bc574 [asan] get rid of the last operator new call in asan rtl
llvm-svn: 160347
2012-07-17 07:20:13 +00:00
Alexey Samsonov 34efb8e9b9 [ASan] Use common flags parsing machinery.
llvm-svn: 159933
2012-07-09 14:36:04 +00:00
Alexey Samsonov 52f946611f [ASan] don't include deleted header
llvm-svn: 158448
2012-06-14 13:15:45 +00:00
Alexey Samsonov 4b1f1031e6 [Sanitizer] factor out GetThreadStackTopAndBottom from ASan runtime to common.
llvm-svn: 158140
2012-06-07 07:13:46 +00:00
Alexey Samsonov 40d5b772e5 [Sanitizer] Switch to common mmap/munmap routines in ASan run-time.
llvm-svn: 158078
2012-06-06 16:15:07 +00:00
Kostya Serebryany 79437fe376 [asan] make tid u32 instead of int
llvm-svn: 158074
2012-06-06 15:06:58 +00:00
Alexey Samsonov 461820ad67 [ASan] Make printf arguments match format strings better.
llvm-svn: 158059
2012-06-06 10:46:00 +00:00
Alexey Samsonov 485d3dc363 Remove file-type tags for .cc files in ASan run-time library
llvm-svn: 157927
2012-06-04 13:50:10 +00:00
Kostya Serebryany 1d35d155fd [asan] more renaming
llvm-svn: 157747
2012-05-31 15:02:07 +00:00
Kostya Serebryany 8d03204204 [asan] more renaming
llvm-svn: 157746
2012-05-31 14:35:53 +00:00
Evgeniy Stepanov d989be1386 Fix GetFrameNameByAddr hitting stale stack guards.
In the current implementation AsanThread::GetFrameNameByAddr scans the
stack for a magic guard value to locate base address of the stack
frame. This is not reliable, especially on ARM, where the code that
stores this magic value has to construct it in a register from two
small intermediates; this register can then end up stored in a random
stack location in the prologue of another function.

With this change, GetFrameNameByAddr scans the shadow memory for the
signature of a left stack redzone instead. It is now possible to
remove the magic from the instrumentation pass for additional
performance gain. We keep it there for now just to make sure the new
algorithm does not fail in some corner case.

llvm-svn: 156710
2012-05-12 12:33:10 +00:00
Alexander Potapenko 08342aa1a1 Introduce the use_sigaltstack flag (off by default), which enables using alternate
per-thread stacks for signal handling. This allows to print more verbose error reports
for stack overflows.

llvm-svn: 154092
2012-04-05 10:54:52 +00:00
Evgeniy Stepanov 823085a8f4 [asan] Support for %z to Printf()
At the moment, asan internal Printf() uses %l modifier for printing
values of size_t and related types. This works, because we control
both the implementation of Printf and all its uses, but can be a
little misleading.

This change adds support for %z to Printf(). All callers that print
sizes and pointers as integers are switched to %zu / %zx.

llvm-svn: 153177
2012-03-21 11:32:46 +00:00
Timur Iskhodzhanov 0f9c9a5332 [ASan] Intercept CreateThread on Windows
llvm-svn: 151366
2012-02-24 15:28:43 +00:00
Kostya Serebryany b5eb5a7a05 [asan] make sure the AsanThread object is destroyed if pthread_exit is called
llvm-svn: 149940
2012-02-07 00:27:15 +00:00
Alexey Samsonov 2d3a67b73b AddressSanitizer: create AsanThreadSummary together with AsanThread (in parent thread)
llvm-svn: 148286
2012-01-17 06:35:31 +00:00
Alexey Samsonov 5d5fa5115a Test commit. Minor cleanup in AsanThread
llvm-svn: 147930
2012-01-11 08:07:51 +00:00
Kostya Serebryany 332923be32 [asan] get rid of the scary TSD destructor code. Now, we store the leaky AsanThreadSummary in TSD and never remove it from there.
llvm-svn: 147910
2012-01-11 02:03:16 +00:00
Kostya Serebryany 5be458ccfd [asan] refactoring: move some common linux/mac code to asan_posix.cc
llvm-svn: 147788
2012-01-09 19:18:27 +00:00
Kostya Serebryany 3f4b9bb4a0 [asan] do not use new/delete for the internal thread structure
llvm-svn: 147674
2012-01-06 19:44:11 +00:00
Kostya Serebryany 78d87d3b4b [asan] move {linux,mac}-specific code from asan_thread.cc to asan_{linux,mac}.cc; also add asan_procmaps.h which I forgot to add on previous commit.
llvm-svn: 147586
2012-01-05 01:07:27 +00:00