Various optimizations to speed up MTasks coarsening (which is the long
pole in the multi-threaded scheduling of very large designs).
The biggest impact ones:
- Use efficient hand written Pairing Heaps for implementing priority
queues and the scoreboard, instead of the old SortByValueMap. This
helps us avoid having to sort a lot of merge candidates that we will
never actually consider and helps a lot in performance.
- Remove unnecessary associative containers and store data structures
(the heap nodes in particular) directly in the object they relate to.
This eliminates a huge amount of lookups and helps a lot in
performance.
- Distribute storage for SiblingMC instances into the LogicMTask
instances, and combine with the sibling maps. This again eliminates
hash table lookups and makes storage structures smaller.
- Remove some now bidirectional edge maps, keep only the forward map.
There are also some other smaller optimizations:
- Replaced more unnecessary dynamic_casts with static_casts
- Templated some functions/classes to reduce the number of static
branches in loops.
- Improves sorting of edges for sibling candidate creation
- Various micro-optimizations here and there
This speeds up MTask coarsening by 3.8x on a large design, which
translates to a 2.5x speedup of the ordering pass in multi-threaded
mode. (Combined with the earlier optimizations, ordering is now 3x
faster.)
Due to the elimination of a lot of the auxiliary data structures, and
ensuring a minimal size for the necessary ones, memory consumption of
the MTask coarsening is also reduced (measured up to 4.4x reduction
though the accuracy of this is low).
The algorithm is identical except for minor alterations of the order
some candidates are added or removed, this can cause perturbation in the
output due to tied scores being broken based on IDs.
__gcov_flush was a private function and was removed from later GCC
versions (at least from 11.2.0, possibly earlier). Replace with the
documented public __gcov_dump.
These have been 'deprecated' for 2 years and are otherwise unused except
for using a temporary placeholder value, which I have inlined with the
default value.
Also remove the now VL_TIME_STR_CONVERT utility function (and
corresponding unit tests), which have no references in any project on
GitHub.
All remaining use of conditional compilation in the tracing
implementation of the run-time library are replaced with the use of
VerilatedModel::traceConfig, and is now done at run-time.
Always fail if adding a model to a trace file that has already executed
a dump. We used to do this before as well, though in a less robust way.
We will be relying on this property more in the future, so improve the
check.
Always build the FST libray with -DFST_WRITER_PARALLEL, iff VL_THREADED.
This supports run-time enablement of the FST writer thread, and has no
measurable performance impact on single threaded tracing but simplifies
the library build.
Note: the actual choice of using the fst writer thread is still compile
time, but can now be made run-time easily.
Step towards a proper run-time library. Reduce the amount of ifdefs in
the implementation of offloaded tracing. There are still a very small
number of ifdefs left, which will need more careful changes in order to
keep user API compatibility.
VCD tracing is now parallelized using the same thread pool as the model.
We achieve this by breaking the top level trace functions into multiple
top level functions (as many as --threads), and after emitting the time
stamp to the VCD file on the main thread, we execute the tracing
functions in parallel on the same thread pool as the model (which we
pass to the trace file during registration), tracing into a secondary
per thread buffer. The main thread will then stitch (memcpy) the buffers
together into the output file.
This makes the `--trace-threads` option redundant with `--trace`, which
now only affects `--trace-fst`. FST tracing uses the previous offloading
scheme.
This obviously helps a lot in VCD tracing performance, and I have seen
better than Amdahl speedup, namely I get 3.9x on XiangShan 4T (2.7x on
OpenTitan 4T).
