| Commit message (Collapse) | Author | Age | Files | Lines |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some projects will get build break because the comipler is confused when
searches for the standard stdio.h. Rename the wrapper file to avoid that.
renamed: src/common/stdio.h -> src/common/stdio_wrapper.h
modified: src/processor/minidump.cc
modified: src/processor/dump_context.cc
modified: src/processor/logging.cc
modified: src/processor/minidump.cc
modified: src/processor/minidump_processor.cc
modified: src/processor/stackwalk_common.cc
modified: src/processor/symbolic_constants_win.cc
R=mark@chromium.org, labath@google.com
Review URL: https://codereview.chromium.org/1864603002 .
Patch from Yunxiao Ma <yxma@google.com>.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This preserves full build ids in minidumps, which are useful for
tracking down the right version of system libraries from Linux
distributions.
The default build id produced by GNU binutils' ld is a 160-bit SHA-1
hash of some parts of the binary, which is exactly 20 bytes:
https://sourceware.org/binutils/docs-2.26/ld/Options.html#index-g_t_002d_002dbuild_002did-292
The bulk of the changes here are to change the signatures of the
FileID methods to use a wasteful_vector instead of raw pointers, since
build ids can be of arbitrary length.
The previous change that added support for this in the processor code
preserved the return value of `Minidump::debug_identifier()` as the
current `GUID+age` treatment for backwards-compatibility, and exposed
the full build id from `Minidump::code_identifier()`, which was
previously stubbed out for Linux dumps. This change keeps the debug ID
in the `dump_syms` output the same to match.
R=mark@chromium.org, thestig@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1688743002 .
|
|
|
|
|
|
| |
Also add the -D__STDC_LIMIT_MACROS definition as we rely on UINTPTR_MAX.
Review URL: https://codereview.chromium.org/1801913002/
|
|
|
|
|
|
|
|
|
|
| |
Based on changes for ARM, ARM64 and X86, the support for
MIPS and MIPS64 is added in microdump.
TEST=microdump_stackwalk ~/microdump-mips32.dmp symbols/
BUG=microdump_stackwalk failing for mips architectures
Review URL: https://codereview.chromium.org/1731923002/
|
|
|
|
|
|
| |
R=ivanpe@chromium.org
Review URL: https://codereview.chromium.org/1849933002 .
|
|
|
|
|
|
|
|
|
|
|
|
| |
to code calling libcurl. This change may be used to build a tool to dump and
upload symbols with multi-thread.
BUG=
R=mmandlis@chromium.org
CC=google-breakpad-dev@googlegroups.com
Review URL: https://codereview.chromium.org/1842113002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Currently EXC_BAD_ACCESS doesn't support EXC_I386_GPFLT as
exception_flags for pretty-printing in the processor, but this happens
for a lot of things:
http://opensource.apple.com/source/xnu/xnu-2050.24.15/osfmk/i386/trap.c
(search for EXC_I386_GPFLT).
And we get a lot of these in the wild:
https://crash-stats.mozilla.com/search/?reason=%3DEXC_BAD_ACCESS+%2F+0x0000000d&cpu_name=amd64&_facets=signature&_facets=address&_columns=date&_columns=signature&_columns=product&_columns=version&_columns=build_id&_columns=platform&_columns=address#crash-reports
This patch makes them show up with a nice name instead of the current
"EXC_BAD_ACCESS / 0x0000000d".
Additionally, this patch fixes some other cases where x86-64 wasn't being handled in the same way as x86, and fixes some x86-specific exception flags to be stringified with I386 in the output.
R=mark@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1833123002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some of the symbols in the stack trace are not found in the .dynsym
section but were located in the full symbol table .symtab section
instead. This was causing some of our stack traces to be incomplete or
point to incorrect function names.
Since we only output function names, there are actually not that many
more symbols located in .symtab that aren't in .dynsym. It is better to
simply output all symbols found so our stack traces are complete.
R=mark@chromium.org, thestig@chromium.org
BUG=561447
Review URL: https://codereview.chromium.org/1824063002 .
Patch from David Yen <dyen@chromium.org>.
|
|
|
|
|
|
|
|
|
|
| |
The auto-conversion from googlecode wiki format apparently wasn't perfect.
This also fixes up some of the links to point to the new Breakpad repo.
R=vapier@chromium.org
BUG=
Review URL: https://codereview.chromium.org/1823583002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
behavior under sized-delete.
The code as it stands allocates a chunk of memory of arbitrary size and places an object into it. It stores a pointer to that object and memory into a list telling the compiler that it is a pointer to a char. When the compiler deletes the objects in the list it thinks that the list contains pointers to chars - not pointers to arbitrarily sized regions of memory.
This is fixing an issue that will reproduces when the following optimization (C++ sized dealocation) is enabled: http://www.open-std.org/jtc1/sc22/wg21/docs/papers/2013/n3536.html
The fix is to explicitly call the non-sized delete operator, and the library code that supports malloc/free/new/delete will figure out the size of the block of memory from the pointer being passed in.
Patch provided by Darryl Gove.
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1788473002 .
|
|
|
|
|
|
|
| |
omap.cc(212): warning C4473: 'fprintf' : not enough arguments passed for format string
BUG=None
Review URL: https://codereview.chromium.org/1776613004
|
|
|
|
|
| |
BUG=https://bugs.chromium.org/p/google-breakpad/issues/detail?id=639
Review URL: https://codereview.chromium.org/1722333002
|
|
|
|
|
|
|
| |
BUG=585534
R=primiano@chromium.org
Review URL: https://codereview.chromium.org/1750033002 .
|
|
|
|
|
|
|
|
|
|
|
| |
The Linux dumpers use absolute paths for shared libraries referenced by
dumps, so they fail to locate them if the crash originated in a chroot.
This CL enables callers to specify a root prefix, which is prepended to
mapping paths before opening them.
BUG=chromium:591792
TEST=make check
Review URL: https://codereview.chromium.org/1761023002/
|
|
|
|
|
|
|
|
| |
Properly handle microdump processing, when the system_log file contains an incomplete microdump section at the top. The processor will process the first complete microdump section.
R=primiano@chromium.org
Review URL: https://codereview.chromium.org/1742843002 .
|
|
|
|
|
|
|
| |
BUG=587536
R=primiano@chromium.org
Review URL: https://codereview.chromium.org/1704243002 .
|
|
|
|
|
|
|
|
|
|
| |
Because tools/windows/symupload/symupload.cc uses `nullptr` (which
requires VS2010), the CLSID comparison is only performed for msdia100.dll
and later. When compiling with an older (or future) CLSID_DiaSource, we
retain the existing behaviour (i.e. fail if CoCreateInstance fails).
R=ivanpe@chromium.org
BUG=https://bugzilla.mozilla.org/show_bug.cgi?id=1236343
|
| |
|
|
|
|
|
|
|
|
|
|
|
| |
There's no need for this binary outside of local tests, so stop
building/installing it by default.
BUG=chromium:579384
TEST=`make check` passe
R=ivanpe@chromium.org
Review URL: https://codereview.chromium.org/1632933002 .
|
|
|
|
|
|
|
| |
BUG=https://code.google.com/p/chromium/issues/detail?id=584174
R=mmandlis@chromium.org
Review URL: https://codereview.chromium.org/1697963002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
If the Linux minidump writer crashes while writing a dump, the dump
might contain some useful information, but the header will be empty
because TypedMDRVA's destructor flushes the data, and the header var
doesn't go out of scope until the end of the `Dump` method. This
fixes that problem by putting the header in a shorter block scope.
We've seen this problem in some Android dumps in the wild, like:
https://crash-stats.mozilla.com/report/index/cef5b777-02d1-43c2-bf40-133ab2160209
R=thestig@chromium.org
BUG=https://bugzilla.mozilla.org/show_bug.cgi?id=1247978
Review URL: https://codereview.chromium.org/1696573003 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Linux make check is failing for mips, mips64, arm, arm64
with error:
"fatal error: mach/arm/vm_types.h: No such file or directory" in case of arm,
"../src/third_party/mac_headers/mach/machine/vm_types.h:37:2: error: #error architecture not supported" in case of mips/mips64
This was partially fixed in https://codereview.chromium.org/1645673002/.
Here excluding src/common/mac/macho_reader_unittest for hosts other than x86/x86-64.
BUG=make check failure for linux mips
TEST=make check pass
Review URL: https://codereview.chromium.org/1692933002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
GPU infromation in the following format:
G GL_VERSION|GL_VENDOR|GL_RENDERER.
The GPU version, vendor and renderer are extracted during microdump parsing and populated in the appropriate fields in the SystemInfo struct.
This is to match the changes introduced in crrev.com/1343713002 and crrev.com/1334473003
BUG=chromium:536769
R=primiano@chromium.org
Review URL: https://codereview.chromium.org/1678463002 .
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts commit cb936a0243c97ae9cd2d4bb19d95dde0421fed6d.
A=dyen@chromium.org
Original Review: https://codereview.chromium.org/1672773002/
R=dyen@chromium.org
Review URL: https://codereview.chromium.org/1688493003 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This patch changes MDCVInfoELF (which is currently unused, apparently
a vestigal bit of code landed as part of Solaris support) into a supported
CodeView format that simply contains a build id as raw bytes.
Modern ELF toolchains support build ids nicely:
https://access.redhat.com/documentation/en-US/Red_Hat_Enterprise_Linux/6/html/Developer_Guide/compiling-build-id.html
It would be useful to have the original build ids of loaded modules in
Linux minidumps, since tools like Fedora's darkserver allow querying by build
id and the current Breakpad code truncates the build id to the size of a GUID,
which loses information:
https://darkserver.fedoraproject.org/
A follow-up patch will change the Linux minidump generation code to produce
MDCVInfoELF in minidumps instead of MDCVInfoPDB70. This patch should be landed
first to ensure that crash processors are able to handle this format before
dumps are generated containing it.
The full build id is exposed as the return value of Minidump::code_identifier(),
which currently just returns "id" for modules in Linux dumps. For
backwards-compatibility, Minidump::debug_identifier() continues to treat
the build id as a GUID, so debug identifiers for existing modules will not
change.
BUG=
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1675413002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The method -[NSURLConnection sendSynchronousRequest:returningResponse:error:]
has been deprecated in 10.11 OS X SDK and 9.0 iOS SDK without replacement. So
emulate a synchronous request by using an asynchronous request and waiting on
a semaphore for the request completion.
BUG=https://bugs.chromium.org/p/google-breakpad/issues/detail?id=675
BUG=569158
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1675243002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The method -[NSString stringByAddingPercentEscapesUsingEncoding:] has been
deprecated with 10.11 OS X SDK and 9.0 iOS SDK. The recommended method is
-[NSString stringByAddingPercentEncodingWithAllowedCharacters:] available
since 10.9 OS X SDK and 7.0 iOS SDK.
Use the new method when available using URLQueryAllowedCharacterSet to get
the same encoded string.
BUG=https://bugs.chromium.org/p/google-breakpad/issues/detail?id=675
BUG=569158
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1680663002 .
|
|
|
|
|
|
|
|
|
|
|
|
| |
The function CFPropertyListCreateFromXMLData is deprecated in favor of
the function CFPropertyListCreateWithData that is available since the
10.6 OS X SDK and 4.0 iOS SDK.
BUG=https://bugs.chromium.org/p/google-breakpad/issues/detail?id=675
BUG=569158
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1678063002 .
|
|
|
|
|
|
|
|
|
|
| |
Adding remaining mips n64 support including stackwalker.
BUG=None
TEST=manually tested on Linux/Android
R=vapier@chromium.org
Review URL: https://codereview.chromium.org/1418453011 .
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Original change (https://codereview.chromium.org/1527363003/) was failing
in CFReadStreamGetBuffer() call, so changed to CFReadStreamRead() to be
more conservative.
Patch provided by Scott Hancher.
BUG=
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1637433003 .
|
|
|
|
|
|
|
|
|
|
|
|
| |
In order to figure out what symbols we need associated to a minidump,
it is useful to be able to dump all the modules the minidump contains.
A=dyen@chromium.org
Original Review: https://codereview.chromium.org/1651593002/
BUG=563716
R=dyen@chromium.org
Review URL: https://codereview.chromium.org/1650713002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This updates the GYP build for the processor component (on windows).
- adds/removes references to files which were added or removed from the
repository
- includes build/common.gypi in the gyp files: needed to correctly
detect the OS (I think, the generated MSVC solutions were broken
without it)
- conditionally compiles code platform-specific code for the given
platform
After this minidump processor nearly compiles with VS2013: the generated
project is correct, but some files still have compilation errors.
Disclaimer: I have not tested the GYP changes on non-windows platform,
as there does not seem to be anyone using it there.
BUG=
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1643633004 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The dump_syms_mac tool only works for the system it is being built for
(it doesn't support running on ELFs for a diff target), and it builds
only for x86 currently.
If you look at the mac header:
src/third_party/mac_headers/mach/machine/vm_types.h
it will #error for non x86/arm systems, and the arm header is not in
our source tree.
Tweak the build so it's only compiled when targetting x86 systems.
BUG=chromium:579384
TEST=`make check` pass
R=ted.mielczarek@gmail.com
Review URL: https://codereview.chromium.org/1645673002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Newer gcc versions default to -Werror=narrowing when using newer C++
standards (which we do). This causes issues when we try to stuff a
value like 0xea into a char -- the value is out of range for signed
char bytes. That's when gcc throws an error:
.../bytereader_unittest.cc: In member function 'virtual void Reader_DW_EH_PE_absptr4_Test::TestBody()':
.../bytereader_unittest.cc:400:55: error: narrowing conversion of '234' from 'int' to 'char' inside { } [-Wnarrowing]
BUG=chromium:579384
TEST=`make check` passes
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1605153004 .
|
|
|
|
|
|
|
|
| |
Should have no real impact on the build. Just pulling in the latest
versions from the latest autoconf/automake versions.
BUG=chromium:579384
TEST=`make && make install` still works
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Some systems provide prebuilt copies of gmock/gtest (such as Chromium
OS). Add a configure flag so they can take advantage of that. This
allows for a smaller checkout as they don't need to include the full
testing/ tree.
BUG=chromium:579384
TEST=`make check` passes w/--enable-system-test-libs
TEST=`make check` passes w/--disable-system-test-libs
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/1638653002 .
|
|
|
|
|
|
|
|
|
|
|
| |
When building with -gsplit-dwarf, the generated dwo files are left behind
even when you `make clean`. Fix that up.
BUG=chromium:579384
TEST=`./configure CXXFLAGS='-O -gsplit-dwarf' && make && make clean` removes dwo files now
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1633893002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The current makefile ends up building ~17 copies of the gtest/gmock
objects -- every test that refers to the cc files directly will have
its own copy. This is because the build doesn't know if CFLAGS and
such have changed between each target (and in some cases, they are).
Create a new libtesting.a target to hold a single copy of these files
and update all of the unittests to link that in. This speeds up the
build a bit especially when you aren't using ccache.
This does mean we can no longer build gtest/gmock with unique flags,
but we haven't wanted that so far, so clearly no one wants that.
BUG=chromium:579384
TEST=`make check` passes
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/1633903002 .
|
|
|
|
|
|
|
|
|
|
|
|
| |
The EXTRA_PROGRAMS knob does not automatically trigger clean up of
targets listed in it. Use CLEANFILES so we make sure `make clean`
will delete the linux_client_unittest_shlib lib.
BUG=chromium:579384
TEST=`make check` passes
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1618593002 .
|
|
|
|
|
|
|
|
|
|
|
| |
Rather than copy & paste the same set of -I flags many times over,
create a single varible to hold it.
BUG=chromium:579384
TEST=`make check` passes
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/1613763002 .
|
|
|
|
|
|
|
|
| |
BUG=chromium:579384
TEST=`make check` passes
R=thestig@chromium.org
Review URL: https://codereview.chromium.org/1607413003 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
This reverts CL https://codereview.chromium.org/1563223004/
This reverts commit 7cc0d8562bf8b20b88cc941ba72593cb7230ecf6.
CL 1563223004 introduces two bugs on iOS.
- Encoding the minidump name with extra percent causing crash server to fail
processing the file.
- Using a released pointer causing random crashes on upload. The
data, resp, err pointers returned in the NSURLSession completion
handler is released at the end of the block. When used later (to get
the crash ID), it causes a crash.
BUG=569158
R=blundell@chromium.org, mark@chromium.org
Review URL: https://codereview.chromium.org/1619603002 .
Patch from Olivier Robin <olivierrobin@chromium.org>.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
The std::getline function always returns its first arg (which is an
iostream object) and cannot return anything else. Thus, testing its
value is pointless, and even leads to build errors w/at least gcc-5
due to gtest ASSERT_TRUE funcs only taking bool types:
.../exploitability_unittest.cc: In member function 'virtual void {anonymous}::ExploitabilityLinuxUtilsTest_DisassembleBytesTest_Test::TestBody()':
.../exploitability_unittest.cc:200:136: error: no matching function for call to 'testing::AssertionResult::AssertionResult(std::basic_istream<char>&)'
In file included from .../breakpad_googletest_includes.h:33:0,
from .../exploitability_unittest.cc:35:
.../gtest.h:262:12: note: candidate: testing::AssertionResult::AssertionResult(bool)
Since we know this never fails, simply drop the ASSERT_TRUE usage.
The next line already checks the content of the buffer we read.
Further on in the file, we hit some signed warnings:
In file included from .../breakpad_googletest_includes.h:33:0,
from .../exploitability_unittest.cc:35:
.../gtest.h: In instantiation of 'testing::AssertionResult testing::internal::CmpHelperEQ(const char*, const char*, const T1&, const T2&) [with T1 = long unsigned int; T2 = int]':
.../gtest.h:1484:23: required from 'static testing::AssertionResult testing::internal::EqHelper<lhs_is_null_literal>::Compare(const char*, const char*, const T1&, const T2&) [with T1 = long unsigned int; T2 = int; bool lhs_is_null_literal = false]'
.../exploitability_unittest.cc:241:289: required from here
.../gtest.h:1448:16: error: comparison between signed and unsigned integer expressions [-Werror=sign-compare]
if (expected == actual) {
This is because we compare the register value (a uint64_t) directly to
an integer constant, and those are signed by default. Stick a U suffix
on them to fix things up.
BUG=chromium:579384
TEST=`make check` passes
R=ivanpe@chromium.org
Review URL: https://codereview.chromium.org/1611763002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Older versions of MSVC don't have a snprintf functions. Some files
were already working around that, but not all of them. Instead of
copying the logic into every file, I centralize it into a new
stdio.h wrapper file and make other files include that.
BUG=
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1602563003 .
Patch from Pavel Labath <labath@google.com>.
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
In C/C++, the result of signed integer overflow is undefined.
The expression "base + size - 1" is parsed as "(base + size) - 1", and
"base + size" can overflow even if "base + (size - 1)" <= INT_MAX.
See http://g/c-compiler-chrome/461JohPKakE/JI3rEBg6FwAJ for more.
BUG=None
TEST='CC=clang CXX=clang++ ./configure && make check'
R=vapier@chromium.org
Review URL: https://codereview.chromium.org/1591793002 .
|
|
|
|
|
|
| |
R=andybons@chromium.org
Review URL: https://codereview.chromium.org/1583163003 .
|
|
|
|
|
|
|
|
|
| |
This CL exports LinuxCoreDumper and ElfCoreDump in the client library. The ARC collector, which is an alternative to core2md optimized for large core dumps, needs these symbols for core dump parsing and conversion to minidump.
BUG=http://b/25773929
TEST=nm src/client/linux/libbreakpad_client.a | grep LinuxCoreDumper
Review URL: https://codereview.chromium.org/1576053002 .
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
windows.h defines exception_code as a macro, which conflicts with our
use of the identifier in exception records. It appears that this
particular include of windows.h is not needed, so instead of undefining
the macro, I simply delete the include. Build tested with MSVC 2013.
BUG=
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1579623004 .
Patch from Pavel Labath <labath@google.com>.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
MSVC does not have the __PTRDIFF_TYPE__ macro defined, so I use the
standard [u]intptr_t types instead. Compilation tested on windows, linux
and mac.
BUG=
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1571293003 .
Patch from Pavel Labath <labath@google.com>.
|
|
|
|
|
|
|
|
|
|
|
|
|
| |
Due to operator precedence, the address was first cast to void*
and then incremented, which resulted in an error on windows, as
sizeof(void) is undefined and MSVC takes this seriously. Changing
the precedence to perform the addition first.
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1570843002 .
Patch from Pavel Labath <labath@google.com>.
|