Partition SuperPMI replay task (#66065)
authorBruce Forstall <brucefo@microsoft.com>
Wed, 2 Mar 2022 23:04:59 +0000 (15:04 -0800)
committerGitHub <noreply@github.com>
Wed, 2 Mar 2022 23:04:59 +0000 (15:04 -0800)
commit83f204eba1fcf541fc4b3dcbd1577d824d213b72
tree8278345126cf0f3a30f005c1a0c9f7cc1fb5918a
parentd21a17f3e8193a071f50e21a46d2b9a980f25a23
Partition SuperPMI replay task (#66065)

* Partition SuperPMI replay task

Split per-platform/architecture work into multiple partitions to increase pipeline
parallelism and reduce overall job time. The partitions are sets of
different JitStressRegs options.

We could create a partition for each JitStressRegs option, but the concern
is that there is potentially a lot of overhead downloading the large set
of MCH files, and we might want to share that overhead between work partitions.

* Add comment
src/coreclr/scripts/superpmi-replay.proj
src/coreclr/scripts/superpmi_replay.py