A real breakpoint makes entries in the breakpoint filter for all offsets at the given line. Various flow control commands will also use the breakpoint filter to add and remove temporary "internal breakpoints" required during the flow operation. Ensure that we never remove a breakpoint filter entry if there was already one there due to a breakpoint (or really any other reason).
No emitted bytecode relies on Continuation being stored in local 0
anymore. Stop using local 0 for this purpose and compute offset
to the Continuation at JIT time. 16 bytes of memory freed.
At this point all locals of Continuation construction wrapper share the
same indices with their respective locals of Continuation body, which
should allow further optimizations.
Flow control commands should not evaluate the conditions of conditional breakpoints when enabling and disabling breakpoints during stepping. Also, all breakpoints, rather than just the first matching breakpoint should be enabled/disabled.
There was a race in the debugger protocol tests which send bad commands to the proxy and expect it to disconnect. If the proxy thread was fast enough it would mark the client as stopped, and we'd skip trying to get a new message after sending the bad command. This means missing one line of output in very rare cases. I've put in a wait to ensure that the client settles in the stopped state before continuing, to remove the race.
A client couldn't break execution during eval. There used to be a lot of barriers to making that right, but I fixed most of them with a previous diff on unifying client-side event loops. Now the only barrier was that a server-side thread processing an interrupt was blocking the signal polling thread by holding a mutex while processing the interrupt. Changed to set a flag to disable polling when starting to process the interrupt (and unsetting it when done), while still synchronizing with the signal polling thread to ensure only one thread is sending the client messages at a time. Added logic to re-enable polling while executing PHP for eval, print, etc. Plumbed the proxy thru to the point where we check the clause on conditional breakpoints, too, since that's the third (and final) place we do this.
break start/end/psp currently always report themselves as unbound. If the client is connected to a sanbox, these should instead be treated as bound. Also, break clear all currently removes breakpoints without running their destructors in the right order, which causes the break point counter to not reset to 1.
In HHVM (and HPHPc before it) we've been piggybacking resources on the
KindOfObject machinery. At the language level, resource is considered to
be a different type than object, and there are a number of differences
in behavior between objects and resources (ex. resources don't allow for
dynamic properties, resources don't work with the clone operator, the
"(object)" cast behaves differently for resources vs. objects, etc).
Piggybacking resources on the KindOfObject machinery has some downsides.
Code that deals with KindOfObject values often needs to check if the value
is a resource and go down a different code path. This makes things harder
to maintain and harder to keep parity with Zend. Also, these extra branches
hurt performance a little, and they make it harder for the JIT to do a good
job in some cases when its generating machine code that operates on objects.
This diff prepares the code base for a new KindOfResource type by adding a
new "Resource" smart pointer type (currently a typedef for the Object smart
pointer type) and it updates the C++ code and the idl files appropriately.
This diff is essentially a cosmetic change and should not impact run time
behavior. In the next diff (part 2) we'll actually add a new KindOfResource
type, detach ResourceData from the ObjectData inheritence hierarchy, and
provide a real implementation for the Resource smart pointer type (instead
of just aliasing the Object smart pointer type).
This abstracts the file format a bit, so that we can
dump to multiple formats. Currently, GraphViz and GML are
supported. GML was added at @edwinsmith's request (and it's much
nicer than GraphViz anyway...)
Adds an extension command to the debugger which provides
heap tracing functionality, and adds the framework needed to get
heap traces. The heaptrace command can dump the current heap to the
debugger session or it can save a GraphViz specification to a file.
I'm hoping to add a backend which can put the graph in GML format,
so it can be explored interactively. I also hope at some point to
see this integrated into FBIDE if we can do that.
Added some test cases. Also now permit caller=>func() type of breakpoints. To make this work, I added a method, getCallingSite, to class InterruptSite and use this to walk the stack when checking if a site matches a break point that has more than one function name.
Updates continuations to allow yielding of a key-value
pair from a generator. Adds bytecode instructions (PackContK,
ContKey) for using the new feature, and adds IR instructions
(ContUpdateIdx, ContIncKey) to help get it down to the metal
(in particular, ContIncKey attempts to keep the current use-cases
as fast as possible).
PackCont opcode is always followed by ContExit so let's join them into ContSuspend. ContResume counterpart may come later (replacing ContNext/Send/Raise/Enter).
The [z]end command in hphpd runs the last manually entered snippet of php in "zend". If you typed 'z' before actually entering any php, it would segfault. Fixed a minor bug in Process::Exec, and present help if none entered. Also clarified that it will simply use your system's default php, and made it so you can override which exe to use with a hphpd config variable. The default is "php", since that is most commonly used.
Add a bit more usage logging for the debugger, and ensure that there is a common field between clients and servers thats easy to search on, i.e., sandbox id.
Because stronger types are better types, and this will make
future refactoring easier. I considered trying to purge the Opcode
type from the codebase too but that would be a much bigger project.
C++11 cleanup (clean up easy enums)
This is for runtime/base/... and ended up touching a lot of files
because it turns out we have a lot of reasonably behaved enums.
This was doing ref(foo), where foo is an Array. This will
create a temporary Variant, call the ref(CVarRef) overload, which
casts the temporary to a non-const RefResult, then a temporary
VRefParam is implicitly constructed from this, which creates a
KindOfRef and points the temporary variant at that. The sort will
operate on this temporary (leading to copy on write, and storing the
new ArrayData* there), so the m_constants field won't reflect the
changes.
The debugger client now accepts feedback from the the debugger server about whether a breakpoint can be hit, absent further loads of files, classes or functions.
There was a similar-but-different event loop used when receiving command results from the server which was close, but not quite right. Unified it with the main event loop to ensure that all error cases are handled properly when we put up a prompt at a nested interrupt, like when hitting a breakpoint during an eval. The event loop is now shared, with a few different "kinds" to control some of the special needs of the loop when executed from a command. Most commands don't cause the server to run more PHP, so they don't change the machine state or cause more interrupts. But some do (Eval and Print) and certainly the top-level loop does, too. Made sure to throw a protocol error if any command causes this to happen when we don't expect it.
Cleanup a lot of hangs with either the debugger client or server in a variety of error conditions, mostly related to communication errors or the client or server exiting unexpectedly. One of the biggest fixes is that all cases where the client was left in a state where Ctrl-C wouldn't work have been fixed.
Remove lots of little snippets of dead code. If you see a function (or small set of functions/fields) deleted then it was actually dead.
I debated whether to keep throwing DebuggerClientExitException on the server, and I decided to keep it. I think it's reasonable that if you've got the server stopped and you quit the debugger that the request gets terminated rather than continuing to run.
I also considered a big change to the way Ctrl-C works, but ended up staying with what was there with just a bit of cleanup. We need to guard against people banging on Ctrl-C, which is a reasonable behavior, and I think it feels pretty reasonable with the updated message.
Finally, added many comments about how this stuff works.
Instead of calling get_systemlib every time the debugger client needs to list source from systemlib.php, use a cached copy of the source string. Only do this if debugging is enabled.
The debugger can only hit breakpoints and list source lines for code lines that have an entry in the byte code repository. In order to make systemlib.php debuggable, it needs an entry. A customized routine has been created for doing this since there is no actual source file available. Also, special case logic is needed to list the source file during debugging.
Small stepping, which is stepping over sub-expressions (kinda), worked but was a little goofy. The mode was set on the client, passed over with control flow commands, placed on the execution context, then retrieved from there and used by those same flow commands. Removed the execution context part of it, since it was useless, and factored grabbing the offsets into the flow cmds where they belong instead of doing it all the time.
The run cmd also had some notion of small stepping, but you'll note it was never sent over the wire. Nuked that, since it never mattered anyway.
We had two similar-but-different functions for getting a notification from the VM about an exception. Cleaned that up by using the proper one for a thrown exception where appropriate, and moving the old one into a hook (like the other VM->debugger hooks) specifically for error messages.
There were multiple issues with flow control when exceptions occur. Fixed these by ditching the reliance on the exception thrown interrupt and introduce an exception handler interrupt, which indicates control is about to pass to a catch clause. This gives us much better insight into how execution is flowing and how we might need to adjust an in-flight stepping operation.
Currently the debugger prints the line where an exception has been thrown, along with the type exception. Then it prints the source listing of the exception. Then it prints a serialization of the exception, which includes a full stack trace. Since the debugger has a command for printing stack traces, the latter bit information seems completely redundant when stopped in the command line client. This diff suppresses the stack trace in that case. It also suppresses redundant diagnostics that get generated during the debugger's attempt to find and print the source code in response to a list command. Finally, the quit command was too eager to let the client die after notifying the proxy, causing the proxy to get a pipe closed exception rather than the quit command, which often allowed the program to carry on running after the client has already quit. The quit command now waits for an acknowledgement from the proxy before shutting down.
Add reasonable behavior for stepping within continuations (generators). Stepping over a yield now does what one would expect. When the generator is driven from a C++ extension like ASIO, the next logical execution point is after the yield statement, and that's where we'll stop now. When driven from PHP, say in a loop calling send(), the next execution point is in fact the call site of send(), so we go there. Stepping off the end of the generator function, goes to the caller of send(), or the caller of the C++ extension. Stepping _out_ of a generator driven by a C++ extension ensures that we go to the caller and not back into the generator again. The logic for both cases is exactly the same. The difference comes from the fact that we don't actually debug C++ extensions.
This also fixes a long-standing problem where breakpoints would interfere with control flow cmds on the same source line. This caused funny behavior, like taking multiple steps to get off of a breakpoint.
Mike Magruder