JIT hooks

There are several hooks in the pypyjit module that may help you with understanding what pypy’s JIT is doing while running your program:

dont_trace_here(next_instr, is_being_profiled, pycode)
get_jitcell_at_key(next_instr, is_being_profiled, pycode)
get_stats_asmmemmgr()

Returns the raw memory currently used by the JIT backend, as a pair (total_memory_allocated, memory_in_use).

residual_call(callable, *args, **keywords)

For testing. Invokes callable(…), but without letting the JIT follow the call.

set_compile_hook(callable, operations=True)

Set a compiling hook that will be called each time a loop is compiled.

The callable will be called with the pypyjit.JitLoopInfo object. Refer to it’s documentation for details.

Note that jit hook is not reentrant. It means that if the code inside the jit hook is itself jitted, it will get compiled, but the jit hook won’t be called for that.

if operations=False, no list of operations will be available. Useful if the hook is supposed to be very lighweight.

set_abort_hook(hook)

Set a hook (callable) that will be called each time there is tracing aborted due to some reason.

The hook will be invoked with the siagnture: hook(jitdriver_name, greenkey, reason, oplist)

Reason is a string, the meaning of other arguments is the same as attributes on JitLoopInfo object

set_trace_too_long_hook(hook)

Set a hook (callable) that will be called each time we abort tracing because the trace is too long.

The hook will be called with the signature: hook(jitdriver_name, greenkey)

enable_debug()

Start recording debugging counters for get_stats_snapshot

Currently disabled.

disable_debug()

Stop recording debugging counters for get_stats_snapshot

Currently disabled.

get_stats_snapshot()

Get the jit status in the specific moment in time. Note that this is eager - the attribute access is not lazy, if you need new stats you need to call this function again. You might want to call enable_debug to get more information. It returns an instance of JitInfoSnapshot

class JitInfoSnapshot

A class describing current snapshot. Usable attributes:

  • counters - internal JIT integer counters
  • counter_times - internal JIT float counters, notably time spent TRACING and in the JIT BACKEND
  • loop_run_times - counters for number of times loops are run, only works when enable_debug is called.
class JitLoopInfo

A class containing information about the compiled loop. Usable attributes:

  • operations - list of operations, if requested
  • jitdriver_name - the name of jitdriver associated with this loop
  • greenkey - a key at which the loop got compiled (e.g. code position, is_being_profiled, pycode tuple for python jitdriver)
  • loop_no - loop cardinal number
  • bridge_no - id of the fail descr
  • type - “entry bridge”, “loop” or “bridge”
  • asmaddr - an address in raw memory where assembler resides
  • asmlen - length of raw memory with assembler associated

Resetting the JIT

releaseall()

Marks all current machine code objects as ready to release. They will be released at the next GC (unless they are currently in use in the stack of one of the threads). Doing pypyjit.releaseall(); gc.collect() is a heavy hammer that forces the JIT roughly back to the state of a newly started PyPy.

set_param(*args, **keywords)

Configure the tunable JIT parameters, paramter names are listed in Jit Help :

  • set_param(name=value, ...) as keyword arguments
  • set_param("name=value,name=value") as a user-supplied string
  • set_param("off") disable the jit
  • set_param("default") restore all defaults