This patch includes multiple changes cleaning up various aspects of the crash
reporter client and exception handler:
* Some Unix-specific code was moved out from the base crashreporter client
code and into the appropriate platform implementation
* Functions used to open files in the crashreporter client now accept C++
`std::ios` flags instead of unreadable booleans
* Useless character conversion routines were removed from the
minidump-analyzer
* Crash annotations are not serialized into a huge string anymore every time
they change. They are all written out individually during an exception.
* `WriteEscapedMozCrashReason()` uses the exception-safe `my_strlen()` instead
of plain `strlen()`
* The Windows-specific DLL-blocklist shutdown was removed from the Linux &
macOS Breakpad callbacks
* The `CrashReporterHost`, `CrashReporterClient` and
`CrashReporterMetadataShmem` classes now take `nsACString` references
instead of `nsCString` ones since they never modify their contents
Differential Revision: https://phabricator.services.mozilla.com/D33267
--HG--
extra : moz-landing-system : lando
Annotation on main process crashes are written to both the .extra file
(for crash submission) and to the event file so that the browser can
detect the crash when restarting even if the crash report files have
been deleted.
This patch factorizes all the code that writes to both files, cutting
down all the duplicate calls, and fixes an issue with the
BreakpadReserveAddress and BreakpadReserveSize annotations which were
not written to the event file.
Differential Revision: https://phabricator.services.mozilla.com/D31247
--HG--
extra : moz-landing-system : lando
Upon a content process crash or hang crash annotations were incrementally
written into the .extra file starting with the exception handler callback and
then in a number of different places before the file was ready for submission.
This had a number of downsides: since the annotations were directly added to
the file it was impossible to tell which ones were already written at a
certain point in time, additionally some were written twice or even thrice.
The code doing the writing would also behave differently depending on the
contents of the file, the parameters passed to it and the contents of global
variables.
This change overhauls the whole process by keeping the annotations into a
temporary per-crash annotation table which is filled with all the required
annotations before being written out in a single pass when they are ready.
The annotations are gathered from the main process annotation table, the
per-process one (held by the CrashReporterHost) and exception-time specific
ones.
The resulting annotations are slightly different than before the patch: first
of all there are no more duplicate entries in the .extra file and secondly all
content/plugin process hangs annotations are properly filtered, before
annotations that were main process-only would leak into them.
Differential Revision: https://phabricator.services.mozilla.com/D31069
--HG--
extra : moz-landing-system : lando
Expose the driver vendor information (applies to Linux only at this
moment) to crash report telemetry and about:support. This will be useful
when debugging issues to know specifically what driver is in use.
Also expose the monitor information for Linux. Part 1 provides an
implementation to get the monitor information on said platform.
Differential Revision: https://phabricator.services.mozilla.com/D29472
Expose the driver vendor information (applies to Linux only at this
moment) to crash report telemetry and about:support. This will be useful
when debugging issues to know specifically what driver is in use.
Also expose the monitor information for Linux. Part 1 provides an
implementation to get the monitor information on said platform.
Differential Revision: https://phabricator.services.mozilla.com/D29472
Expose the driver vendor information (applies to Linux only at this
moment) to crash report telemetry and about:support. This will be useful
when debugging issues to know specifically what driver is in use.
Also expose the monitor information for Linux. Part 1 provides an
implementation to get the monitor information on said platform.
Differential Revision: https://phabricator.services.mozilla.com/D29472
Moved changes to the non-forked part of breakpad living under
toolkit/crashreporter/google-breakpad into separate patches that are applied
by update-breakpad.sh when synchronizing with upstream breakpad. Because we
landed the commits directly to the sources every time we called
update-breakpad.sh those changes would be reverted.
Differential Revision: https://phabricator.services.mozilla.com/D27682
--HG--
extra : moz-landing-system : lando
This includes improvements to the Linux exception handler that will provide
crash addresse for a number of signals by reading the NT_SIGINFO structure and
quites warnings in a number of files. This also removes an unused header.
Differential Revision: https://phabricator.services.mozilla.com/D26488
--HG--
extra : moz-landing-system : lando
After replacing precise line information from .debug_line with coarse
line information from DW_AT_call_{file,line}, it's very likely that
adjacent line records actually refer to identical file and line
numbers. Such adjacent records are not really useful and take up more
space than they should in the symbol file. We might as well merge them
and save ourselves some space.
Differential Revision: https://phabricator.services.mozilla.com/D25473
--HG--
extra : moz-landing-system : lando
DW_TAG_subprogram DIEs sometimes have child DW_TAG_lexical_block DIEs
which in turn contain child DW_TAG_inlined_subroutine DIEs that we woud
like to look at. If we skip the DW_TAG_inlined_subroutine DIEs, we miss
important information. We therefore need to look through the
DW_TAG_lexical_block DIEs to find the DIEs that we are interested in.
Depends on D25471
Differential Revision: https://phabricator.services.mozilla.com/D25472
--HG--
extra : moz-landing-system : lando
We record the file and line that these subroutines were inlined from.
We'll use that information to provide more coarse-grained line
information in the next patch.
Depends on D25469
Differential Revision: https://phabricator.services.mozilla.com/D25470
--HG--
extra : moz-landing-system : lando
The DW_AT_call_file attributes that we eventually want to parse from
DW_TAG_inlined_subroutine DIEs refer to the file name table stored in
the .debug_line section. To resolve those DW_AT_call_file attributes,
we need access to that table after parsing of the appropriate
.debug_line bits is done. This patch adds support for extracting that
information from the .debug_line parsing process.
Differential Revision: https://phabricator.services.mozilla.com/D25469
--HG--
extra : moz-landing-system : lando
I changed DelegatedQueryInterface and CallMethod to be non-static
methods rather than taking an explicit |self| parameter.
There is already a method nsXPCWrappedJS::CallMethod() with the same
signature, but it is a shim, so I inlined it into the version in
XPCWrappedJSClass.cpp.
I also fixed up a few comments that mention nsXPCWrappedJSClass.
The new comments starting with "We now need to enter the realm" were
written by Boris, and are a little more explicit so they are easier to
understand.
I renamed DebugDump() to DebugDumpInterfaceInfo() to be more
informative.
Differential Revision: https://phabricator.services.mozilla.com/D26422
--HG--
extra : moz-landing-system : lando
I changed DelegatedQueryInterface and CallMethod to be non-static
methods rather than taking an explicit |self| parameter.
I did a tiny bit of cleanup in the nsIXPConnectJSObjectHolder case of
DelegatedQueryInterface().
There is already a method nsXPCWrappedJS::CallMethod() with the same
signature, but it is a shim, so I inlined it into the version in
XPCWrappedJSClass.cpp.
I also fixed up a few comments that mention nsXPCWrappedJSClass.
The new comments starting with "We now need to enter the realm" were
written by Boris, and are a little more explicit so they are easier to
understand.
I renamed DebugDump() to DebugDumpInterfaceInfo() to be more
informative.
Differential Revision: https://phabricator.services.mozilla.com/D26422
--HG--
extra : moz-landing-system : lando
Module::AddFunction assumes that any time we add a function, it is
preferable to use that instead of the corresponding external
symbol (which could be anything). The former show up in Breakpad symbol
files as FUNC lines, and the latter as PUBLIC lines.
For Module::AddFunction to be effective, we have to parse all the
external symbols for a module first, and then discover all the actual
functions. But the Linux symbol dumping code does the reverse: it first
parses all the DWARF information (including .debug_info, which adds any
relevant functions) and then parses the ELF symbol table. This ordering
means that we wind up emitting PUBLIC lines for which corresponding FUNC
lines already exist. These duplicate PUBLIC lines take up roughly 10%
of the size of a libxul symbol file (~30MB), and are completely unnecessary.
The fix is simple: we should reverse the order in which we parse ELF
symbols and DWARF debug information.
Differential Revision: https://phabricator.services.mozilla.com/D25116
--HG--
extra : moz-landing-system : lando
The definitions can't be entirely removed yet because NSS still needs them.
Differential Revision: https://phabricator.services.mozilla.com/D23454
--HG--
extra : moz-landing-system : lando
Disable tests as they fail constantly on xpcshell-8 on windows10-aarch64.
Bug 1536221 is filed for investigating why the crashreporter tests still fail.
Differential Revision: https://phabricator.services.mozilla.com/D23940
--HG--
extra : moz-landing-system : lando
