Skip to content
Merged
Changes from all commits
Commits
File filter

Filter by extension

Filter by extension

Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
4 changes: 3 additions & 1 deletion src/coreclr/scripts/superpmi_aspnet.py
Original file line number Diff line number Diff line change
Expand Up @@ -174,7 +174,9 @@ def build_and_run(coreclr_args):

# note tricks to get one element tuples

runtime_options_list = [("Dummy=0",), ("TieredCompilation=0", ), ("TieredPGO=1", "TC_QuickJitForLoops=1"), ("TieredPGO=1", "TC_QuickJitForLoops=1", "ReadyToRun=0")]
runtime_options_list = [("Dummy=0",), ("TieredCompilation=0", ), ("TieredPGO=1", "TC_QuickJitForLoops=1"), ("TieredPGO=1", "TC_QuickJitForLoops=1", "ReadyToRun=0"),
Copy link
Member

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

should first items have TC_OnStackReplacement=0 as you're planning to enable it by default?

Copy link
Member Author

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

I suppose so, though with the current defaults (~50,000 hits to patchpoints before a transition) very few methods will end up with OSR. In practice it won't make much difference.

The bigger impact will be from changing the default for QJFL.

("TC_QuickJitForLoops=1", "ReadyToRun=0", "TC_OnStackReplacement=1", "OSR_HitLimit=0", "TC_OnStackReplacement_InitialCounter=0"),
("TieredPGO=1", "TC_QuickJitForLoops=1", "ReadyToRun=0", "TC_OnStackReplacement=1", "OSR_HitLimit=0", "TC_OnStackReplacement_InitialCounter=100")]

# runtime_options_list = [("TieredCompilation=0", )]

Expand Down