These changes were made in the upstream version of crashpad without
being contributed back to crashpad.
Bug: crashpad:271
Change-Id: I60f6dfd206191e65bac41978a7c88d06b8c3cee9
Reviewed-on: https://chromium-review.googlesource.com/c/1389238
Commit-Queue: Vlad Tsyrklevich <vtsyrklevich@chromium.org>
Reviewed-by: Mark Mentovai <mark@chromium.org>
Crashpad has many tests that crash intentionally. Some of these are
gtest death tests, and others arrange for intentional crashes to test
Crashpad’s own crash-catching logic. On macOS, all of the gtest death
tests and some of the other intentional crashes were being logged by
ReportCrash, the system’s crash reporter. Since these reports
corresponded to intentional crashes, they were never useful, and served
only to clutter ~/Library/Logs/DiagnosticReports.
Since Crashpad is adept at handling exceptions on its own, this
introduces the “exception swallowing server”,
crashpad_exception_swallower, which is a Mach exception server that
implements a no-op exception handler routine for all exceptions
received. The exception swallowing server is established as the task
handler for EXC_CRASH and EXC_CORPSE_NOTIFY exceptions during gtest
death tests invoked by {ASSERT,EXPECT}_DEATH_{CHECK,CRASH}, and for all
child processes invoked by the Multiprocess test infrastructure. The
exception swallowing server is not in effect at other times, so
unexpected crashes in test code can still be handled by ReportCrash or
another crash reporter.
With this change in place, no new reports are generated in the
user-level ~/Library/Logs/DiagnosticReports or the system’s
/Library/Logs/DiagnosticReports during a run of Crashpad’s full test
suite on macOS.
Bug: crashpad:33
Change-Id: I13891853a7e25accc30da21fa7ea8bd7d1f3bd2f
Reviewed-on: https://chromium-review.googlesource.com/777859
Commit-Queue: Mark Mentovai <mark@chromium.org>
Reviewed-by: Robert Sesek <rsesek@chromium.org>
The design for running all Crashpad unit tests on Chromium’s try- and
buildbots involves pulling all tests into a single monolithic
crashpad_tests executable. Many Crashpad tests base the name of their
child executables or modules on the name of the main test executable.
Since the main test executable will have a different name in the
in-Chromium build, knowledge of the test executable name (referred to as
“module” here) needs to be added to the tests themselves.
This introduces TestPaths::BuildArtifact(), which allows the module name
to be specified. For Crashpad’s standalone build, the module name is
verified against the main test executable’s name.
TestPaths::BuildArtifact() can also locate paths in the alternate 32-bit
output directory for 64-bit Windows tests, taking on the responsibility
for what the new (5e9ed4cb9f69) TestPaths::Output32BitDirectory(), now
obsolete, did.
Bug: chromium:779790
Change-Id: I64c4a2190b6319e487c999812a7cfc512a75a700
Reviewed-on: https://chromium-review.googlesource.com/747536
Reviewed-by: Scott Graham <scottmg@chromium.org>
Nothing currently directs the handler to read these Annotation objects
from the target process, so they will not be read by Crashpad nor appear
in the minidump.
Bug: crashpad:192
Change-Id: I8ebabb4f5c77c5620b0d8e5036c3185eecfa4646
Reviewed-on: https://chromium-review.googlesource.com/717236
Commit-Queue: Robert Sesek <rsesek@chromium.org>
Reviewed-by: Mark Mentovai <mark@chromium.org>
This uses “static” at function scope to avoid making local copies, even
in cases where the compiler can’t see that the local copy is
unnecessary. “constexpr” adds additional safety in that it prevents
global state from being initialized from any runtime dependencies, which
would be undesirable.
At namespace scope, “constexpr” is also used where appropriate.
For the most part, this was a mechanical transformation for things
matching '(^| )const [^=]*\['.
Similar transformations could be applied to non-arrays in some cases,
but there’s limited practical impact in most non-array cases relative to
arrays, there are far more use sites, and much more manual intervention
would be required.
Change-Id: I3513b739ee8b0be026f8285475cddc5f9cc81152
Reviewed-on: https://chromium-review.googlesource.com/583997
Commit-Queue: Mark Mentovai <mark@chromium.org>
Reviewed-by: Leonard Mosescu <mosescu@chromium.org>
gtest used to require (expected, actual) ordering for arguments to
EXPECT_EQ and ASSERT_EQ, and in failed test assertions would identify
each side as “expected” or “actual.” Tests in Crashpad adhered to this
traditional ordering. After a gtest change in February 2016, it is now
agnostic with respect to the order of these arguments.
This change mechanically updates all uses of these macros to (actual,
expected) by reversing them. This provides consistency with our use of
the logging CHECK_EQ and DCHECK_EQ macros, and makes for better
readability by ordinary native speakers. The rough (but working!)
conversion tool is
https://chromium-review.googlesource.com/c/466727/1/rewrite_expectassert_eq.py,
and “git cl format” cleaned up its output.
EXPECT_NE and ASSERT_NE never had a preferred ordering. gtest never made
a judgment that one side or the other needed to provide an “unexpected”
value. Consequently, some code used (unexpected, actual) while other
code used (actual, unexpected). For consistency with the new EXPECT_EQ
and ASSERT_EQ usage, as well as consistency with CHECK_NE and DCHECK_NE,
this change also updates these use sites to (actual, unexpected) where
one side can be called “unexpected” as, for example, std::string::npos
can be. Unfortunately, this portion was a manual conversion.
References:
https://github.com/google/googletest/blob/master/googletest/docs/Primer.md#binary-comparison77d6b17338https://github.com/google/googletest/pull/713
Change-Id: I978fef7c94183b8b1ef63f12f5ab4d6693626be3
Reviewed-on: https://chromium-review.googlesource.com/466727
Reviewed-by: Scott Graham <scottmg@chromium.org>
This supports the “double handler” or “double handler with low
probability” models from https://crashpad.chromium.org/bug/143.
For crashpad_handler to be become its own client, it needs access to its
own executable path to pass to CrashpadClient::StartHandler(). This was
formerly available in the test-only test::Paths::Executable(). Bring
that function’s implementation to the non-test Paths::Executable() in
util/misc, and rename test::Paths to test::TestPaths to avoid future
confusion.
test::TestPaths must still be used to access TestDataRoot(), which does
not make any sense to non-test code.
test::TestPaths::Executable() is retained for use by tests, which most
likely prefer the fatal semantics of that function. Paths::Executable()
is not fatal because for the purposes of implementing the double
handler, a failure to locate the executable path (which may happen on
some systems in deeply-nested directory hierarchies) shouldn’t cause the
initial crashpad_handler to abort, even if it does prevent a second
crashpad_handler from being started.
Bug: crashpad:143
Test: crashpad_util_test Paths.*, crashpad_test_test TestPaths.*
Change-Id: I9f75bf61839ce51e33c9f7c0d7031cebead6a156
Reviewed-on: https://chromium-review.googlesource.com/466346
Reviewed-by: Scott Graham <scottmg@chromium.org>
Commit-Queue: Mark Mentovai <mark@chromium.org>
ReadFile() attempted to continue reading after a short read. In most
cases, this is fine. However, ReadFile() would keep trying to fill a
partially-filled buffer until experiencing a 0-length read(), signaling
end-of-file. For certain weird file descriptors like terminal input, EOF
is an ephemeral condition, and attempting to read beyond EOF doesn’t
actually return 0 (EOF) provided that they remain open, it will block
waiting for more input. Consequently, ReadFile() and anything based on
ReadFile() had an undocumented and quirky interface, which was that any
short read that it returned (not an underlying short read) actually
indicated EOF.
This facet of ReadFile() was unexpected, so it’s being removed. The new
behavior is that ReadFile() will return an underlying short read. The
behavior of FileReaderInterface::Read() is updated in accordance with
this change.
Upon experiencing a short read, the caller can determine the best
action. Most callers were already prepared for this behavior. Outside of
util/file, only crashpad_database_util properly implemented EOF
detection according to previous semantics, and adapting it to new
semantics is trivial.
Callers who require an exact-length read can use the new
ReadFileExactly(), or the newly renamed LoggingReadFileExactly() or
CheckedReadFileExactly(). These functions will retry following a short
read. The renamed functions were previously called LoggingReadFile() and
CheckedReadFile(), but those names implied that they were simply
wrapping ReadFile(), which is not the case. They wrapped ReadFile() and
further, insisted on a full read. Since ReadFile()’s semantics are now
changing but these functions’ are not, they’re now even more distinct
from ReadFile(), and must be renamed to avoid confusion.
Test: *
Change-Id: I06b77e0d6ad8719bd2eb67dab93a8740542dd908
Reviewed-on: https://chromium-review.googlesource.com/456676
Reviewed-by: Robert Sesek <rsesek@chromium.org>
Apple has responded to their bug 29079442 with a resolution stating that
these are not corpse ports but task ports that have changed after
execve(), as part of the large task port and execve() strategy rewrite
from 10.12.1. The comments being replaced were written before we had
10.12.1 source code. Now that we can see what’s going on, revise the
comments, and re-enable the task port check for the non-execve() test
variants.
https://openradar.appspot.com/29079442https://googleprojectzero.blogspot.com/2016/10/taskt-considered-harmful.html
Bug: crashpad:137
Test: crashpad_snapshot_test MachOImageAnnotationsReader.CrashDyld
Change-Id: I463637816085f4165b92b85a5b98bfeddcdf4094
Reviewed-on: https://chromium-review.googlesource.com/451120
Reviewed-by: Robert Sesek <rsesek@chromium.org>
Commit-Queue: Mark Mentovai <mark@chromium.org>
Use “macOS” as the generic unversioned name of the operating system in
comments. For version-specific references, use Mac OS X through 10.6, OS
X from 10.7 through 10.11, and macOS for 10.12.
Change-Id: I1ebee64fbf79200bc799d4a351725dd73257b54d
Reviewed-on: https://chromium-review.googlesource.com/408269
Reviewed-by: Robert Sesek <rsesek@chromium.org>
crashpad_snapshot_test MachOImageAnnotationsReader.CrashDyld was failing
on 10.12.1. In 10.12, dyld’s intentional crashes come through
abort_with_payload(). In 10.12.1, it appears that the task port sent
along with abort_with_payload() crashes is now a corpse port, which has
a different port name than the task port that it originated from.
https://openradar.appspot.com/29079442
TEST=crashpad_snapshot_test MachOImageAnnotationsReader.CrashDyld
BUG=crashpad:137
Change-Id: I43f89c0f595dd5614fc910fa1f19f21ddf0a7c4d
Reviewed-on: https://chromium-review.googlesource.com/407087
Reviewed-by: Robert Sesek <rsesek@chromium.org>
In 10.12, dyld calls abort_with_payload() on fatal error from
dyld::halt(). In previous 10.12 betas, abort_with_payload() caused the
process to appear to terminate as exit(1). This was weird, so I filed
https://openradar.appspot.com/26894758. In 10.12db4 16A270f, Apple seems
to have fixed this bug. abort_with_payload() as used by dyld now causes
the process to appear to terminate as abort() as I had requested.
A Crashpad test that assures Crashpad’s ability to catch dyld crashes
needs its expectations updated with each change to a process’ apparent
termination code. It’s updated to expect SIGABRT on 10.12 or later. No
concessions are made for previous 10.12 betas or their buggy exit(1)
behavior. Nobody should be running any 10.12 beta prior to 10.12db4
16A270f now or at any point in the future.
This undoes (redoes) 335ef494677f.
BUG=crashpad:120
TEST=crashpad_snapshot_test MachOImageAnnotationsReader.CrashDyld
Change-Id: I13b330ac83fc9b33907ac172d35983974b8910f0
Reviewed-on: https://chromium-review.googlesource.com/365920
Reviewed-by: Robert Sesek <rsesek@chromium.org>
exit(1) is a weird code for this, so I filed
https://openradar.appspot.com/26894758.
This doesn’t completely fix bug crashpad:121 unless both
crashpad_snapshot_test and crashpad_snapshot_test_no_op are signed with
the same Developer ID certificate. I’m hoping to get some action on
https://openradar.appspot.com/26902656, which will enable a complete fix
for this bug in unsigned developer builds. It would be unusual to have
to sign test executables.
BUG=crashpad:120,crashpad:121
TEST=crashpad_snapshot_test MachOImageAnnotationsReader.CrashDyld
Change-Id: I54fdfaa9178029b91ea3cbc12f2760dfa5124858
Reviewed-on: https://chromium-review.googlesource.com/355260
Reviewed-by: Robert Sesek <rsesek@chromium.org>
This was done in Chromium’s local copy of Crashpad in 562827afb599. This
change is similar to that one, except more care was taken to avoid
including headers from a .cc or _test.cc when already included by the
associated .h. Rather than using <stddef.h> for size_t, Crashpad has
always used <sys/types.h>, so that’s used here as well.
This updates mini_chromium to 8a2363f486e3a0dc562a68884832d06d28d38dcc,
which removes base/basictypes.h.
e128dcf10122 Remove base/move.h; use std::move() instead of Pass()
8a2363f486e3 Move basictypes.h to macros.h
R=avi@chromium.org
Review URL: https://codereview.chromium.org/1566713002 .
CrashReportExceptionHandler::CatchMachException() must always set a
valid new_state. Failing to do so appears to trigger corpse generation
on OS X 10.11. This is addressed by calling ExcServerCopyState().
Previously, this was not done for exceptions forwarded to the user
ReportCrash, under the apparent mistaken assumption that ReportCrash
would do it. However, ReportCrash is given copies of out-parameters like
new_state to explicitly prevent it from influencing Crashpad’s returned
state.
ExcServerSuccessfulReturnValue() must not return MACH_RCV_PORT_DIED for
an EXC_CRASH handler on OS X 10.11. This appears to trigger corpse
generation. This is addressed by always returning KERN_SUCCESS from
EXC_CRASH handlers on OS X 10.11.
This also adds generic EXC_CORPSE_NOTIFY support throughout Crashpad.
The crashpad_handler does not listen for this exception type, but it is
now possible to work with this exception type using tools like
exception_port_tool and catch_exception_tool.
BUG=crashpad:48
TEST=Crashes handled by crashpad_handler do not result in the generation
of reports in the root /Library/Logs/DiagnosticReports.
R=kerrnel@chromium.org, rsesek@chromium.org
Review URL: https://codereview.chromium.org/1305893010 .
The system’s crashreporter_annotations_t structure was always present
as version 4 since Mac OS X 10.7. In OS X 10.11, it is now present as
version 5. It has also grown from 56 to 64 bytes per otool examination
of CoreFoundation’s __DATA,__crash_info section. The extra 8 bytes are
presumed to be a new field at the end of the structure, although this
is not confirmed.
The existing MachOImageAnnotationsReader.CrashAbort test only validated
that the “message” field in crashreporter_annotations_t was recovered
correctly, but
MachOImageAnnotationsReader::ReadCrashReporterClientAnnotations() also
recovers the “message2” field. A new test,
MachOImageAnnotationsReader.CrashModuleInitialization, is added to
ensure that the “messgae2” field can be recovered properly.
This change will resolve warnings such as:
[pid:tid:yyyymmdd,hhmmss.uuuuuu:WARNING
mach_o_image_annotations_reader.cc:82] unexpected crash info version 5
in
/System/Library/Frameworks/CoreFoundation.framework/Versions/A/CoreFoundation
BUG=crashpad:40
TEST=crashpad_snapshot_test MachOImageAnnotationsReader.CrashAbort,
MachOImageAnnotationsReader.CrashModuleInitialization
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1277513003 .
OS X 10.11 introduces System Integrity Protection. One facet of that
forbids code injection into system executables. A Crashpad test checks
that information can be recovered from dyld in early-launch crashes by
requesting dyld load a nonexistent library with DYLD_INSERT_LIBRARIES.
The executable was meaningless but a system-provided executable,
/usr/bin/true, was used for convenience.
This test hung on OS X 10.11 because DYLD_INSERT_LIBRARIES was ignored
for the system executable, and no crash occurred. The test waited for a
crash that would never come.
A custom no-op executable, crashpad_snapshot_test_no_op, is provided as
an executable that does work with DYLD_INSERT_LIBRARIES.
BUG=crashpad:41
TEST=crashpad_snapshot_test MachOImageAnnotationsReader.CrashDyld
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1276553005 .
After 9e79ea1da719, it no longer makes sense for crashpad_util_test_lib
to “hide” in util/util_test.gyp. All of util/test is moved to its own
top-level directory, test, which all other test code is allowed to
depend on. test, too, is allowed to depend on all other non-test code.
In a future change, when crashpad_util_test_lib gains a dependency on
crashpad_client, it won’t look so weird for something in util (even
though it’s in util/test) to depend on something in client, because the
thing that needs to depend on client will live in test, not util.
BUG=crashpad:33
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1051533002
nullptr ProcessReader::Module.
Prior to 64b87325b9de, the alignment problem meant that the Module for
dyld was looking at the wrong address instead of dyld’s correct load
address when a 32-bit process attempted to examine a crashing 64-bit
process. This resulted in a crash during the
MachOImageAnnotationsReader.CrashDyld test.
ProcessReader::Module pointers are permitted to be nullptr. This allows
minimal module data (its name) to be preserved even when no sense can be
made of the module based on its load address. The producer,
ProcessReader::InitializeModules(), and the non-test consumer,
ModuleSnapshotMac::Initialize(), both accept this correctly. The
producer’s documentation is updated to call this out. The ProcessReader
test is also updated to tolerate this case without crashing by adding
assertions.
TEST=snapshot_test MachOImageAnnotationsReader.*, ProcessReader.*
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/989713002
- Various "FD" to "Handle"
- Existing Multiprocess implementation moves to _posix.
- Stub implementation for _win.
At the moment, multiprocess_exec_win.cc contains implementations of both
Multiprocess methods and MultiprocessExec functions. This will need more
work in the future, but reflects the idea that all tests should be in
terms of MultiprocessExec eventually.
Currently, this works sufficiently to have util_test succeed (including
multiprocess_exec_test, and the recently ported HTTPTransport tests.)
R=mark@chromium.org
BUG=crashpad:1, crashpad:7
Review URL: https://codereview.chromium.org/880763002
Rename fd_io to file_io, and ReadFD to ReadFile, etc.
file_io.cc is the higher level versions that call the basic ReadFile/WriteFile
and then file_io_posix.cc and file_io_win.cc are the implementations of
those functions.
The Windows path is as yet untested, lacking the ability to link the test binary.
R=cpu@chromium.org, mark@chromium.org
BUG=crashpad:1
Review URL: https://codereview.chromium.org/811823003
MachMessageServer::Run()’s distinct |nonblocking| parameter is removed.
The information it formerly conveyed is now implied by the |timeout_ms|
parameter, which can accept two special values,
kMachMessageTimeoutNonblocking and kMachMessageTimeoutWaitIndefinitely.
TEST=client_test, snapshot_test, util_test
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/777993002
UniversalMachExcServer provided both an interface and an implementation,
contrary to the other classes in the exc_server_variants family. This
was mostly done for reasons of economy in an already-large class family.
Unfortunately, this decision meant that it was impossible for other code
to use UniversalMachExcServer, which required that CatchMachException()
be implemented, and also extend another class without violating the
style guide’s prohibition of multiple implementation inheritance. This
became a problem in a lot of test code, which extended MachMultiprocess
and UniversalMachExcServer.
UniversalMachExcServer is now given its own nested Interface class,
which is a pure interface. All users of UniversalMachExcServer are
changed from “is-a” UniversalMachExcServer to “has-a”
UniversalMachExcServer and “is-a” UniversalMachExcServer::Interface.
TEST=client_test, snapshot_test, util_test
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/775943005
Previously, MachMessageServer::Run() only provided two strategies for
dealing with large messages, indicated by mach_msg() returning
MACH_RCV_TOO_LARGE: the receive buffer could be reallocated and the
message received, or the entire function could return MACH_RCV_TOO_LARGE
to the caller. There are situations where an intermediate behavior might
be desirable. This intermediate behavior would allow the function to
continue waiting for another message without returning an error to the
caller or attempting to receive the large message. This is desirable
when dealing with fixed-sized messages and a receiver that might be sent
messages by unknown, possibly-malicious callers. This can happen when
the corresponding send right is published with the bootstrap server, for
example.
Existing users continue to request their existing behavior, typically
receiving an error when encountering a large message.
catch_exception_tool will use the new “ignore” behavior when running in
persistent mode.
TEST=util_test MachMessageServer.*
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/756803002
