Fix some typos and references to member functions that didn't make the
final cut.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@381 4c0a9323-5329-0410-9bdc-e9ce6186880e
src/linux/common/module.h defines a new class, google_breakpad::Module,
that can represent the contents of a breakpad symbol file. Module::Write
writes a well-formed symbol file to the given stream.
src/linux/common/dump_symbols.cc can now lose its symbol-file-writing
code, and change DumpStabsHandler to populate a Module object, rather
than the old SymbolInfo/SourceFileInfo/... collection of types.
The code to compute function and line sizes, even in the absence of
reliable size data in STABS, is moved into a new Finalize method of
DumpStabsHandler, which is responsible for completing the Module's
contents.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@380 4c0a9323-5329-0410-9bdc-e9ce6186880e
With this patch, dump_symbols.cc no longer knows about the details of
the STABS debugging format; that is handled by the StabsReader class.
dump_symbols.cc provides a subclass of StabsHandler that builds
dump_symbols' own representation of the data.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@378 4c0a9323-5329-0410-9bdc-e9ce6186880e
Because the actual N_FUN strings in the .stabstr section contain type
information after the mangled name, representing this information
using a pointer into .stabstr, while efficient with memory, makes the
FuncInfo data structure STABS-specific: one must know the details of a
STABS N_FUN string's syntax to interpret FuncInfo::name. This patch
removes this STABS dependency from the data structure, and moves us
closer to having an appropriate structure for representing unified
STABS and DWARF data.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@375 4c0a9323-5329-0410-9bdc-e9ce6186880e
In STABS, if one function's line number information contains an N_SOL
entry to switch to a new source file, then the next function's line
data should pick up in the same source file where the prior function
left off. However, the Linux dumper restarts each function in the
compilation unit's main source file. This patch fixes that, so that
the output attributes the lines in subsequent functions to the correct
source files.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@373 4c0a9323-5329-0410-9bdc-e9ce6186880e
Let LineInfo structures point directly to their SourceLineInfo
structures, rather than holding the index of the file's name in the
.stabstr section in the early phases, and then later the holding
source_id of the file.
This is another step in the process of moving STABS-specific values
out of the types that represent the breakpad symbol data. When we're
done, the non-STABS structures will be something that we can populate
with both STABS and DWARF data --- or at least it will be more easily
replaced with such.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@371 4c0a9323-5329-0410-9bdc-e9ce6186880e
std::vector::erase() invalidates the iterator, so we need
to advance the iterator by using the return value of erase().
R=nealsid
A=wtc
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@370 4c0a9323-5329-0410-9bdc-e9ce6186880e
STABS information introduces a compilation unit with an N_SO entry
whose address is the start address of the file and whose string is the
name of the compilation unit's main source file. However, STABS
entries can only hold one address, so STABS indicates the compilation
unit's ending address with an N_SO entry whose name is empty.
Currently, the dumper's data structures simply create SourceFileInfo
structures with empty names for these end-of-unit N_SO entries. We
want to remove STABS-specific characteristics from these structures so
that we can replace them with an input-format-independent structure.
This moves end-of-compilation-unit addresses out of the symbol table
structure, and into their own list of boundary addresses.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@369 4c0a9323-5329-0410-9bdc-e9ce6186880e
Use a list of pointers to SourceFileInfo structures, not a list of the
structures themselves. This is preparation for a subsequent patch
which makes the data structures less STABS-specific.
This patch introduces a memory leak. If an included file is
referenced only by line entries for functions that LoadFuncSymbols
elected to omit from the func_info list, then its SourceFileInfo
structure is leaked when we destroy the name_to_file map. This leak
is fixed in a subsequent patch by letting the map of files by name own
the file objects.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@368 4c0a9323-5329-0410-9bdc-e9ce6186880e
Replace the sorted lists of files and functions with an array of
boundary addresses. This replaces CompareAddress with the default
comparison, and SortByAddress and NextAddress with the stock STL sort
and upper_bound algorithms, losing ~50 lines of code.
a=jimblandy
r=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@367 4c0a9323-5329-0410-9bdc-e9ce6186880e
In NextAddress, check both the file list and the function list for the
nearest boundary. Don't assume that, if we find any bounding entry in
the function list, that must be the nearest thing.
A=jimblandy
R=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@365 4c0a9323-5329-0410-9bdc-e9ce6186880e
The current arrangement would produce needless warnings if
WriteSymbolFile were ever used twice in the same program invocation.
Even if it weren't wrong, it's unnecessary, and local non-const static
variables require extra care when reading to be sure of their effect.
A=jimblandy
R=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@363 4c0a9323-5329-0410-9bdc-e9ce6186880e
With this patch, the time required to generate Breakpad symbols for
Firefox's libxul.so on a MacBook Pro 3,1 drops from 32s to 2s.
I verified that this patch had no effect on the output of dump_syms
when applied to firefox-bin and its libraries when built with -gstabs+.
A=jimblandy
R=nealsid
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@362 4c0a9323-5329-0410-9bdc-e9ce6186880e
- Made localization for UI entirely string-based, with flexible layout based on the size of the strings inserted.
- Made the request for an email address optional.
- Fixed a bug that would prevent comments or email from being collected if the text field were still focused.
- Refactored askUserPermissionToSend.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@335 4c0a9323-5329-0410-9bdc-e9ce6186880e
This upload fixes five issues:
1) Preston's email was hardcoded in the xib :-(
2) Changed from xib to NIB to facilitate Tiger building
3) Changed the logs location to be user specifiable by BreakpadMinidumpLocation
key, or ~/Library/Breakpad/<BREAKPAD_PRODUCT> by default
4) Fixed GTM Defines problem in order to build on Tiger
5) Also set CFBundleIcon in the sender program correctly, and updated plist, and
renamed ReporterIcons to crash_report_sendER.ICNS. However the rietveld upload
script doesn't appear to pick up renamed files correctly, so that file doesn't
show up in the patch upload.
Also various comments were updated for accuracy.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@323 4c0a9323-5329-0410-9bdc-e9ce6186880e
Written by Ginn Chen & Eagle.Lu@
R=nealsid (although I don't have a Solaris machine to build on, & these changes look localized to Sun-only builds)
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@314 4c0a9323-5329-0410-9bdc-e9ce6186880e
The method of calculating a binary ID using the LC_ID command isn't compatible with non-default build processes, most Mac consumers
use LC_UUID anyway but for those that don't, MD5 is a better choice
R=nealsid
W=Ted.Mielczarek
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@313 4c0a9323-5329-0410-9bdc-e9ce6186880e
the graceful code inside a #ifdef.
The problem is:
If ExceptionHandler is created and destroyed in DllMain, then the previous change to
remove the call to TerminateThread will lead to a deadlock. This is because inside
DllMain the loader lock is acquired, and the previous change waits for the handler
thread to exit in the destructor, that is with the loader lock acquired. But the
handler thread cannot finish until it gets the loader lock to call DllMain for
THREAD_DETACH.
With this change, we add conditional compilation so that clients that want to
avoid the call to terminate thread can do it by defining the appropriate preprocessor
variable.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@309 4c0a9323-5329-0410-9bdc-e9ce6186880e
Modified dump_syms to detect dSYM bundles or a binary with DWARF data appropriately, and convert data from DWARF reader to dump_syms native structures
R=danny.berlin (original writer of DWARF code)
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@286 4c0a9323-5329-0410-9bdc-e9ce6186880e
only in case of in-process dump generation; otherwise the dump processing code of
breakpad doesn't identify the crashing thread properly.
I am checking this in to-be-reviewed since it is needed urgently in one of the
products at Google.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@276 4c0a9323-5329-0410-9bdc-e9ce6186880e
details (second half of the post).
Changing PROCESS_ALL_ACCESS to GENERIC_ALL in ClientInfo class when
opening the client process handle on the OOP server side.
A minor change to pipe name in the test app also.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@274 4c0a9323-5329-0410-9bdc-e9ce6186880e
make the code compatible for teams that still use VS 2003 to build.
Fix a bug in client ClientInfo class that returned CustomClientInfo struct containing
a pointer in the client process address space instead of pointer to in-process data
structure.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@271 4c0a9323-5329-0410-9bdc-e9ce6186880e
scenarios that the OOP server can use in whatever way it wants to.
Fix a bug in CrashGenerationserver where CreateNamedPipe failure was not checked correctly.
TODO in near future: Add a custom stream to minidump files for the custom information.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@267 4c0a9323-5329-0410-9bdc-e9ce6186880e
tests(ok, just one) for the private copy of nlist that I checked in
last week, plus lay some of the ground work for collecting code
coverage numbers. Both of these are accomplished by leveraging the
built-in facilities of Xcode & Developer Tools(namely, CPlusTest and
gcov integration; however, eventually I will also add a way to get
lcov results from the gcov results). I also:
- renamed breakpad_nlist_64.c to breakpad_nlist_64.cc to be more
consistent(even though it's not C++ code it still only gets called by
C++ code so I don't have to deal with extern "C" constructs).
- I created a new target (minidump_tests) that has a "Coverage"
configuration with the appropriate GCC flags turned on. It is only
compiled in 64-bit configurations and has 10.5 as a minimum
deployment target as well as uses the 10.5 SDK.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@260 4c0a9323-5329-0410-9bdc-e9ce6186880e
64-bit builds of Leopard. I ported the code over myself and will
check it into the Breakpad tree until we decide there's a better
longer-term solution. If you want to diff the changes, the Apple
source is in libc/gen/nlist.c(I used 498 from 10.5.2).
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@259 4c0a9323-5329-0410-9bdc-e9ce6186880e
security attributes on the pipe. This is useful when the process
hosting the crash server wants to let processes from other users to
connect to it and generate dumps for them.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@258 4c0a9323-5329-0410-9bdc-e9ce6186880e
I waited for the style reviewer to approve all my code so hopefully there
should be no more style related check-ins.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@256 4c0a9323-5329-0410-9bdc-e9ce6186880e
Have sent an email for code review already but this check-in is urgent
to fix potential build breaks.
It contains the following changes:
- Change project settings for exception_handler.vcproj so that it links in
crash_generation.lib so that the applications using exception_handler.lib
do not have to link to crash_generation.lib. Make these changes for all 4
build configurations.
- Change the test app project settings to not link to crash_generation.lib
and only link to exception_handler.lib; again for all four configurations.
These changes help the applications that are currently linking to exception_handler.lib
in that they don't have to link to one more lib to get out-of-process dump generation
functionality. In fact, even if they do not use out-of-process dump generation functionality,
after those changes, the apps will need to link with crash_generation.lib anyway in order to
build correctly. This change will eliminate that need.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@248 4c0a9323-5329-0410-9bdc-e9ce6186880e
added in my last check-in for out-of-process dump generation. My last check-in broke the breakpad_client.sln
build for non debug builds since for the new project that I added (crash_generation.vcproj) to that solution,
I always only changed project properties for debug build. In fact, I didn't have the static-crt build configurations
for the new project either. Similar problems existed for the test application I added, windows/test/crash_generation_app).
This changelist contains changes to 2 solution files, 2 vcproj files and 1 .cc file. The fixes to .cc file are to avoid
warnings on variables used only in debug modes (in asserts) when building in release modes.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@247 4c0a9323-5329-0410-9bdc-e9ce6186880e
Implement out-of-process dump generation for Windows platform.
Details:
- Created a lib, crash_generation.lib, that implements the out-of-process dump generation protocol.
- The lib code is under client/windows/crash_generation folder and is organized in the following way:
- CrashGenerationServer class (crash_generation_server.h/.cc) implements the server side of
the protocol.
- CrashGenerationClient class (crash_generation_client.h/.cc) implements the client side of
the protocol.
- MinidumpGenerator class (minidump_generator.h/.cc) serves as an abstractino for generating
dump files using Windows APIs, coming up with new file names by creating GUIDs, etc.
- ProtocolMessage class (ipc_protocol.h) represents the message format between the client and server
for pipe IPC.
- Server allows one client at a time on the pipe in the current implementation.
- ReadMe.txt explains the state machine the server uses to serve clients.
- ExceptionHandler is modified and a new constructor is added that allows specifying the pipe name. If the
pipe name is NULL, the behavior is backward compatible - in-process dump generation is done as before. If
the pipe name is specified, out-of-process dump generation registration is attempted. If that fails, the
behavior is again backward compatible.
- If out-of-process registration succeeds, all write dump requests, direct or indirect, are directed to
crash server process that served the registration request. NOTE that the explicit dump requests made by
calling the static method of ExceptionHandler are not directed to theserver.
- client/windows/tests/crash_generation_app implements a simple Win32 GUI application to help test the
out-of-process dump generation client and server. Typical use of the app is to start one instance, click
Server --> Start and then start the other instance. The other instance will register with the first
instance automatically at start-up. Then the second instance can be used to request various typoes of
dump requests by using options under the Client menu.
git-svn-id: http://google-breakpad.googlecode.com/svn/trunk@244 4c0a9323-5329-0410-9bdc-e9ce6186880e
