MEMORY_BASIC_INFORMATION64 specifies an alignment of 16, but the
standard allocator used by containers doesn't honor this. Although 16
is the default alignment size used on Windows for x86_64, it's not for
32-bit x86. clang assumed that the alignment of the structure was as
declared, and used an SSE load sequence that required this alignment.
AlignedAllocator is a replacement for std::allocator that allows the
alignment to be specified. AlignedVector is an std::vector<> that uses
AlignedAllocator instead of std::allocator.
BUG=chromium:564691
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1498133002 .
Don't call GetFileVersionInfo(), which calls LoadLibrary() to be able to
access the module's resources. Loading modules from the crashy process
into the handler process can cause trouble. The Crashpad handler
definitely doesn't want to run arbitrary modules' module initializer
code.
Since the VS_FIXEDFILEINFO needed is already in memory in the remote
process' address space, just access it from there.
BUG=crashpad:78
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1475023004 .
This unifies several things that used a 16-character random string, and
a few other users of random identifiers where it also made sense to use
a 16-character random string.
TEST=crashpad_util_test RandomString.RandomString
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1451793002 .
The bug and linked code review has more of the history, but we’ve been
tempted to remove the loop outright a couple of times already before
realizing that it serves an important purpose. Hopefully this comment
will protect our future selves from going on the same fool’s errand.
BUG=crashpad:75
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1427643010 .
This is better because now end_to_end_test.py fails immediately with
[1180:9020:20151106,145204.830:ERROR registration_protocol_win.cc:39] CreateFile: The system cannot find the file specified. (0x2)
R=mark@chromium.org
BUG=crashpad:75
Review URL: https://codereview.chromium.org/1409693011 .
In 2eeaa3ac549a, I added a check to make sure that the expected CONOUT$
handle was found. Its omission seemed to be unintentional. The tests
passed for me on Windows 10, but failed on the bots. I can reproduce
the failures locally on Windows 7.
Doing the inheritance test with a file other than CONOUT$ fixes the
immediate problem, but we should find out why this CONOUT$ handle isn't
showing up in the handles list on Windows 7, fix it, and add back a
test.
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1428753009 .
This consolidates all of the twisted casts and comments that discuss how
HANDLEs are really only 32 bits wide even in 64-bit processes on 64-bit
operating systems into a single location.
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1422503015 .
Allowing the client to create its own pipe name string caused a race
between client and server. Instead, in this mode, the server now creates
the pipe name along with a pipe, and returns it to its client via a
--handshake-handle. This guarantees that by the time the client gets the
pipe name, the server has already created it.
Ephemeral mode is now implied by --handshake-handle. The --persistent
option is gone. --persistent mode is enabled when using --pipe-name.
BUG=crashpad:69
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1432563003 .
I considered writing the CodeView records to the minidump, but I didn't
find a ton of docs and debugging is only lightly supported (e.g.
http://www.debuginfo.com/articles/gendebuginfo.html#debuggersandformats
and it doesn't attempt to load at all on more recent Visual Studios).
As we won't be generating symbols in this format, and we don't expect to
have symbols for any weird modules that get injected into us in the
wild, it seems like we don't lose anything by just ignoring them.
R=mark@chromium.org
BUG=crashpad:47
Review URL: https://codereview.chromium.org/1430773003 .
Each routine in this implementation returns MIG_BAD_ID. These routines
may be overridden.
Most things that implement NotifyServer::Interface will only need to
implement one of the interface routines. Since another user of
NotifyServer will be added soon, it makes sense to provide a default
no-op implementation rather than forcing everyone to write the same
no-op boilerplate repeatedly.
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1414413006 .
Previously, crashpad_handler made its own receive right, and transferred
a corresponding send right to its client. There are two advantages to
making the receive right in the client:
- It is possible to monitor the receive right for a port-destroyed
notificaiton in the client, allowing the handler to be restarted if
it dies.
- For the future run-from-launchd mode (bug crashpad:25), the handler
will obtain its receive right from the bootstrap server instead of
making its own. Having the handler get its receive right from
different sources allows more code to be shared than if it were to
sometimes get a receive right and sometimes make a receive right and
transfer a send right.
This includes a restructuring in crashpad_client_mac.cc that will make
it easier to give it an option to restart crashpad_handler if it dies.
The handler starting logic should all behave the same as before.
BUG=crashpad:68
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1409073013 .
This allows better code sharing in crashpad_handler’s main(). It doesn’t
look like much of an improvement now, but a separate change will cause
the Mac ExceptionHandlerServer() to be constructed with an argument. It
will be beneficial for Mac and Windows to be able to share the Run()
call.
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1402333004 .
The intended use is to flip the client-server relationship in
CrashpadClient so that the initial client (parent process) furnishes the
handler process with a receive right. The parent can optionally receive
a port-destroyed notification allowing it to restart the handler if it
exits prematurely.
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1408473002 .
Fixes two incorrect usages of ssize_t/off_t being implicitly converted
to bool. As such, I think it's worth the cost of the additional !! on
BOOL returning Win32 functions.
R=mark@chromium.org
Review URL: https://codereview.chromium.org/1408123006 .
In https://codereview.chromium.org/1411523006, the Mach port scopers are
becoming better ScopedGenerics and are losing the type conversion
operators in the process. This is needed to adapt to that change. get()
is ugly, but being explicit about conversion isn’t a bad thing, and
these scopers will gain functionality such as Pass() as part of the
switch.
As a bonus, some would-be uses of get() to check for valid port rights
are becoming a more descriptive is_valid().
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1405273002 .
This new test from 7de04b02f85d was failing on Windows 10. I started by
adding the hint, which produced “CreateFileMapping: Access is denied.
(0x5)â€. Switching the “Global\†to “Local\†fixes the test for me.
TEST=crashpad_util_test ProcessInfo.Handles
R=scottmg@chromium.org
Review URL: https://codereview.chromium.org/1407993003 .
I thought I had confirmed that this still allocated and ignored the flag
on older OSs, but I must have not had the PLOG active yet? I'm not sure
what I did. (I might try to blame VMware as it has an annoying habit of
caching old binaries when you use it's "Shared Folders" feature to point
at the dev machine's build dir.)
I confirmed that it does work on Win8 and Win10 but doesn't on Win XP
and Win 7.
R=mark@chromium.org
BUG=crashpad:52
Review URL: https://codereview.chromium.org/1405243002 .
Capture the memory for the loader lock (can be inspected by !cs), as
well as all locks that were created with .DebugInfo which can be viewed
with !locks.
e.g.
0:000> !cs ntdll!LdrpLoaderLock
-----------------------------------------
Critical section = 0x778d6410 (ntdll!LdrpLoaderLock+0x0)
DebugInfo = 0x778d6b6c
NOT LOCKED
LockSemaphore = 0x0
SpinCount = 0x04000000
0:000> !locks -v
CritSec ntdll!RtlpProcessHeapsListLock+0 at 778d7620
LockCount NOT LOCKED
RecursionCount 0
OwningThread 0
EntryCount 0
ContentionCount 0
CritSec +7a0248 at 007a0248
LockCount NOT LOCKED
RecursionCount 0
OwningThread 0
EntryCount 0
ContentionCount 0
CritSec crashy_program!g_critical_section_with_debug_info+0 at 01342c48
LockCount NOT LOCKED
RecursionCount 0
OwningThread 0
EntryCount 0
ContentionCount 0
CritSec crashy_program!crashpad::`anonymous namespace'::g_test_critical_section+0 at 01342be0
WaiterWoken No
LockCount 0
RecursionCount 1
OwningThread 34b8
EntryCount 0
ContentionCount 0
*** Locked
Scanned 4 critical sections
R=mark@chromium.org
BUG=crashpad:52
Review URL: https://codereview.chromium.org/1392093003 .
This doesn’t really provide compatibility, it just ignores the
deprecation warning for +[NSURLConnection
sendSynchronousRequest:returningResponse:error:].
The suggested replacement, NSURLSession, was new in 10.9, and this code
needs to run on 10.6, so it’s not usable here, at least not without a
runtime check.
BUG=crashpad:65
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1395673002 .
We already use all the shared constants for page protection and type,
so rather than making various incompatible structures, just use
the MEMORY_BASIC_INFORMATION64 one directly, so that it can be directly
used.
R=mark@chromium.org
BUG=crashpad:20, crashpad:46
Review URL: https://codereview.chromium.org/1375313005 .
ExceptionPorts::GetExceptionPorts() returned a
std::vector<ExceptionPorts::ExceptionHandler>, which contained send
rights to Mach ports. The interface required callers to assume ownership
of each send right contained within the vector. This was cumbersome and
error-prone, and despite the care taken in Crashpad, port right leaks
did occur:
- SimulateCrash() didn’t make any attempt to release these resources at
all.
- Neither did crashpad_util_test ExceptionPorts.HostExceptionPorts,
which also reused a vector.
This replaces the vector with the interface-compatible (as far as
necessary) ExceptionPorts::ExceptionHandlerVector, which deallocates
collected port rights on destruction or clear().
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1381023007 .
This wraps bootstrap_check_in() in BootstrapCheckIn(), and
bootstrap_look_up() in BootstrapLookUp(). The wrappers make it more
difficult to accidentally leak a returned right. They’re easier to use,
encapsulating common error checking and logging, simplifying all call
sites.
TEST=crashpad_util_test MachExtensions.BootstrapCheckInAndLookUp
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1383283003 .
Chrome’s relauncher process needs a way to sever ties with the
crashpad_handler instance running from the disk image in order to cause
that instance to exit so that the disk image may be unmounted. This new
function is otherwise not thought to be interesting, and its use is not
recommended.
This comes with a small refactoring to create a
SystemCrashReporterHandler() function, and a fix for a minor port leak
in CrashReportExceptionHandler::CatchMachException().
BUG=chromium:538373
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1375573005 .
Sadly this code did not survive a collision with the real world. In
probing for the environment block there's a MEM_COMMIT region followed
directly by a MEM_RESERVE region (past the end of the environment
block).
Update region checker to correctly treat MEM_RESERVE as inaccessible.
R=mark@chromium.org
BUG=crashpad:20, crashpad:46, crashpad:59
Review URL: https://codereview.chromium.org/1370063005 .
On Win10, VirtualQueryEx supports querying the x64 part of WOW64
processes. However, on lower OSs it errors past 2/3G. There's no direct
way to retrieve to maximum memory address for processes other than
yourself, but fortunately, VirtualQueryEx sets a distinct error code
when `lpAddress` exceeds the maximum accessible address, so we can just
terminate successfully in that case.
R=mark@chromium.org
BUG=crashpad:20, crashpad:46
Review URL: https://codereview.chromium.org/1376353002 .
Windows requires the connection to the handler to do anything, so it
can't really be implemented or tested without CrashpadClient and the
connection machinery.
R=mark@chromium.org
BUG=crashpad:53
Review URL: https://codereview.chromium.org/1356383002 .
This makes the basics of !peb work in windbg, however, pointed-to things
are not yet retrieved. For full functionality, a variety of pointers in
the PEB also needs to be walked and captured.
e.g.
Previously:
0:000> .ecxr
eax=00000007 ebx=7e383000 ecx=c3f9a943 edx=00000000 esi=006d62d0 edi=003c9280
eip=00384828 esp=005bf634 ebp=005bf638 iopl=0 nv up ei pl zr na pe nc
cs=0023 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010246
crashy_program!crashpad::`anonymous namespace'::SomeCrashyFunction+0x28:
00384828 c7002a000000 mov dword ptr [eax],2Ah ds:002b:00000007=????????
0:000> !peb
PEB at 7e383000
error 1 InitTypeRead( nt!_PEB at 7e383000)...
Now:
0:000> .ecxr
eax=00000007 ebx=7f958000 ecx=02102f4d edx=00000000 esi=00e162d0 edi=01389280
eip=01344828 esp=00c2fb64 ebp=00c2fb68 iopl=0 nv up ei pl zr na pe nc
cs=0023 ss=002b ds=002b es=002b fs=0053 gs=002b efl=00010246
crashy_program!crashpad::`anonymous namespace'::SomeCrashyFunction+0x28:
01344828 c7002a000000 mov dword ptr [eax],2Ah ds:002b:00000007=????????
0:000> !peb
PEB at 7f958000
InheritedAddressSpace: No
ReadImageFileExecOptions: No
BeingDebugged: No
ImageBaseAddress: 01340000
Ldr 77ec8b40
*** unable to read Ldr table at 77ec8b40
SubSystemData: 00000000
ProcessHeap: 00e10000
ProcessParameters: 00e114e0
CurrentDirectory: '< Name not readable >'
WindowTitle: '< Name not readable >'
ImageFile: '< Name not readable >'
CommandLine: '< Name not readable >'
DllPath: '< Name not readable >'
Environment: 00000000
Unable to read Environment string.
R=mark@chromium.org
BUG=crashpad:46
Review URL: https://codereview.chromium.org/1364053002 .
Removes the bitness-specific targets in favour of pulling binaries from
the other build directory. This is to avoid the added complexity of
duplicating all the targets for the x86 in x64 build.
Overall, mostly templatizing more functions to support the
wow64-flavoured structures. The only additional functionality required
is reading the x86 TEB that's chained from the x64 TEB when running
as WOW64.
The crashing child test was switched to a manual CreateProcess because
it needs to launch a binary other than itself.
R=mark@chromium.org
BUG=crashpad:50
Review URL: https://codereview.chromium.org/1349313003 .
A few function implementations that were missing, various switches
for functions/functionality that didn't exist on XP, and far too long
figuring out what exactly was wrong with SYSTEM_PROCESS_INFORMATION
on x86 (the "alignment_for_x86" fields).
R=mark@chromium.org
BUG=crashpad:1, crashpad:50, chromium:531663
Review URL: https://codereview.chromium.org/1336823002 .
The pipe handle was being leaked on connections (oops!). On XP this
resulted in the next test's CreateNamedPipe to fail, because the
previous one still existed (because all handles were not closed). More
recent OSs are more forgiving so I got away with the buggy code.
R=mark@chromium.org
BUG=crashpad:50
Review URL: https://codereview.chromium.org/1337953003 .
PROCESS_ALL_ACCESS was changed in later SDKs and the newer value fails
when run on XP with ERROR_ACCESS_DENIED. Use the old value to maintain
compatibility with XP.
R=mark@chromium.org
BUG=crashpad:50
Review URL: https://codereview.chromium.org/1337133002 .
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 .
This replaces the registration server, and adds dispatch to a delegate
on crash requests.
(As you are already aware) we went around in circles on trying to come
up with a slightly-too-fancy threading design. All of them seemed to
have problems when it comes to out of order events, and orderly
shutdown, so I've gone back to something not-too-fancy.
Two named pipe instances (that clients connect to) are created. These
are used only for registration (which should take <1ms), so 2 should be
sufficient to avoid any waits. When a client registers, we duplicate
an event to it, which is used to signal when it wants a dump taken.
The server registers threadpool waits on that event, and also on the
process handle (which will be signalled when the client process exits).
These requests (in particular the taking of the dump) are serviced
on the threadpool, which avoids us needing to manage those threads,
but still allows parallelism in taking dumps. On process termination,
we use an IO Completion Port to post a message back to the main thread
to request cleanup. This complexity is necessary so that we can
unregister the threadpool waits without being on the threadpool, which
we need to do synchronously so that we can be sure that no further
callbacks will execute (and expect to have the client data around
still).
In a followup, I will readd support for DumpWithoutCrashing -- I don't
think it will be too difficult now that we have an orderly way to
clean up client records in the server.
R=cpu@chromium.org, mark@chromium.org, jschuh@chromium.org
BUG=crashpad:1,crashpad:45
Review URL: https://codereview.chromium.org/1301853002 .
Calling std::vector<>::operator[]() with an out-of-range index argument
is undefined behavior. In two cases, Crashpad used &v[0] in situations
where it was known that the address would not be used. These calls were
wrapped in conditions guarding against vector emptiness.
While s[0] is valid on an empty string, in two cases, Crashpad used
&s[0] as an argument to a system call that would be a no-op. These calls
were wrapped in similar conditions to avoid the system call.
The two uses of vector with undefined behavior were caught by the
following tests in crashpad_snapshot_test with
UndefinedBehaviorSanitizer:
[ RUN ] CrashpadInfoClientOptions.OneModule
/Users/mark/compilatorium/llvm.build/bin/../include/c++/v1/vector:1493:12:
runtime error: reference binding to null pointer of type
'crashpad::process_types::section'
[ OK ] CrashpadInfoClientOptions.OneModule (72 ms)
[ RUN ] ProcessSnapshotMinidump.Empty
/Users/mark/compilatorium/llvm.build/bin/../include/c++/v1/vector:1493:12:
runtime error: reference binding to null pointer of type
'MINIDUMP_DIRECTORY'
[ OK ] ProcessSnapshotMinidump.Empty (1 ms)
The Crashpad codebase was audited by searching for resize() calls and
analyzing how resized strings and vectors are used.
TEST=*
BUG=
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1283243004 .
Found by -fsanitize=undefined:
[ RUN ] Launchd.CFPropertyToLaunchData_FloatingPoint
../../../util/mac/launchd_test.mm:82:33: runtime error: value
1.79769e+308 is outside the range of representable values of type
'float'
[ OK ] Launchd.CFPropertyToLaunchData_FloatingPoint (2 ms)
TEST=crashpad_util_test Launchd.CFPropertyToLaunchData_FloatingPoint
BUG=
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1302843004 .
While not strictly asan-related, this bug was found while running tests
under asan. Evidently, strings are pooled differently in that build
configuration.
TEST=crashpad_util_test ExceptionPorts.*
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1291573004 .
HTTPTransport.Upload33k failed on Windows due to WinHTTP timing out. The
test server, http_transport_test_server.py, writes the entire request to
a stdout pipe, to be received by crashpad_util_test. crashpad_util_test
is also the HTTP client, and it does not attempt to read from this pipe
until the HTTP transaction is complete. http_transport_test_server.py
must not write to stdout until the transaction is complete, otherwise,
there is a risk of deadlock if the pipe buffer fills up. The new
Upload33k test sends a large request, which was filling up the pipe
buffer on Windows.
This also adds an Upload33k_LengthUnknown test variant to exercise a
large POST when the length is not known ahead of time. This more closely
matches how Crashpad crash uploads are done on OS X.
TEST=crashpad_util_test HTTPTransport.*
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1286173007 .
CFStream’s CFReadStreamGetBuffer() calls the Read() callback without
initializing at_eof. The callback function is responsible for setting it
on any successful read operation. See 10.10.2 CF-1152.14/CFStream.c.
By chance, at_eof seems to always have an initial value of false on
x86_64, but true on 32-bit x86. Crashpad’s Read() callback assumed that
the initial value was always false. The discrepancy caused truncation
and possibly hangs when a 32-bit process attempted to upload a request
body larger than 32kB, the buffer size used by NSMutableURLRequest or
something between it and CFReadStream.
A new test with more than 32kB of data is added.
As discussed in:
https://groups.google.com/a/chromium.org/d/topic/crashpad-dev/Vz--qMZJRPU
TEST=crashpad_util_test HTTPTransport.Upload33k
BUG=
R=rsesek@chromium.org
Review URL: https://codereview.chromium.org/1304433004 .
This test was added in https://codereview.chromium.org/1052813002. It
was previously checking the timestamp from in-memory module traversal
vs. the disk mtime. This is flaky (of course) because it depends on
the linker writing the header and closing the file during the same time
quantum. So the bots occasionally failed with:
[ RUN ] ProcessInfo.Self
e:\b\build\slave\chromium_win_dbg\build\crashpad\util\win\process_info_test.cc(86): error: Value of: GetTimestampForModule(GetModuleHandleW(nullptr))
Actual: 1431650338
Expected: modules[0].timestamp
Which is: 1431650337
Instead, use imagehlp to pull the timestamp out of the header so that
it matches the header value that will be the in-memory timestamp.
R=cpu@chromium.orgTBR=mark@chromium.org
Review URL: https://codereview.chromium.org/1139103003
Retrieve context and save to thread context. NtQueryInformationThread
is no longer required (right now?) because to retrieve the CONTEXT, the
thread needs to be Suspend/ResumeThread'd anyway, and the return value
of SuspendThread is the previous SuspendCount.
I haven't handle the x86 case yet -- that would ideally be via
Wow64GetThreadContext (I think) but unfortunately that's Vista+, so I'll
likely need to to a bit of fiddling to get that sorted out. (It's actually
likely going to be NtQueryInformationThread again, but one thing at a
time for now.)
R=cpu@chromium.org, rsesek@chromium.orgTBR=mark@chromium.org
BUG=crashpad:1
Review URL: https://codereview.chromium.org/1133203002
The next big piece of functionality in snapshot. There's a bit more
grubbing around in the NT internals than would be nice, and it has
made me start to question the value avoiding MinidumpWriteDump. But
this seems to extract most of the data we need (I haven't pulled
the cpu context yet, but I hope that won't be too hard.)
R=mark@chromium.org
BUG=crashpad:1
Review URL: https://codereview.chromium.org/1131473005
