diff options
author | Alan Wu <XrXr@users.noreply.github.com> | 2019-07-30 21:36:05 -0400 |
---|---|---|
committer | 卜部昌平 <shyouhei@ruby-lang.org> | 2019-10-24 18:03:42 +0900 |
commit | 89e7997622038f82115f34dbb4ea382e02bed163 (patch) | |
tree | 993a5f6fb17418381e835be1fd51093dc620148a /mjit.c | |
parent | 38e931fa2ceac6d922f3eabedb8f35f211de0bdb (diff) |
Combine call info and cache to speed up method invocation
To perform a regular method call, the VM needs two structs,
`rb_call_info` and `rb_call_cache`. At the moment, we allocate these two
structures in separate buffers. In the worst case, the CPU needs to read
4 cache lines to complete a method call. Putting the two structures
together reduces the maximum number of cache line reads to 2.
Combining the structures also saves 8 bytes per call site as the current
layout uses separate two pointers for the call info and the call cache.
This saves about 2 MiB on Discourse.
This change improves the Optcarrot benchmark at least 3%. For more
details, see attached bugs.ruby-lang.org ticket.
Complications:
- A new instruction attribute `comptime_sp_inc` is introduced to
calculate SP increase at compile time without using call caches. At
compile time, a `TS_CALLDATA` operand points to a call info struct, but
at runtime, the same operand points to a call data struct. Instruction
that explicitly define `sp_inc` also need to define `comptime_sp_inc`.
- MJIT code for copying call cache becomes slightly more complicated.
- This changes the bytecode format, which might break existing tools.
[Misc #16258]
Notes
Notes:
Merged: https://github.com/ruby/ruby/pull/2564
Diffstat (limited to 'mjit.c')
-rw-r--r-- | mjit.c | 11 |
1 files changed, 10 insertions, 1 deletions
@@ -46,7 +46,16 @@ mjit_copy_job_handler(void *data) const struct rb_iseq_constant_body *body = job->iseq->body; if (job->cc_entries) { - memcpy(job->cc_entries, body->cc_entries, sizeof(struct rb_call_cache) * (body->ci_size + body->ci_kw_size)); + unsigned int i; + struct rb_call_cache *sink = job->cc_entries; + const struct rb_call_data *calls = body->call_data; + const struct rb_kwarg_call_data *kw_calls = (struct rb_kwarg_call_data *)&body->call_data[body->ci_size]; + for (i = 0; i < body->ci_size; i++) { + *sink++ = calls[i].cc; + } + for (i = 0; i < body->ci_kw_size; i++) { + *sink++ = kw_calls[i].cc; + } } if (job->is_entries) { memcpy(job->is_entries, body->is_entries, sizeof(union iseq_inline_storage_entry) * body->is_size); |